paper_id
stringlengths
10
10
title
stringlengths
15
153
paper_url
stringlengths
42
42
authors
listlengths
1
71
type
stringclasses
3 values
primary_area
stringclasses
0 values
abstract
large_stringlengths
422
2.2k
keywords
listlengths
0
0
TL;DR
large_stringclasses
0 values
submission_number
int64
2
10.2k
arxiv_id
stringlengths
10
10
arxiv_id_source
stringclasses
2 values
1sRuv4cnuZ
Multi-Track Message Passing: Tackling Oversmoothing and Oversquashing in Graph Learning via Preventing Heterophily Mixing
https://openreview.net/forum?id=1sRuv4cnuZ
[ "Hongbin Pei", "Yu Li", "Huiqi Deng", "Jingxin Hai", "Pinghui Wang", "Jie Ma", "Jing Tao", "Yuheng Xiong", "Xiaohong Guan" ]
Spotlight
null
The advancement toward deeper graph neural networks is currently obscured by two inherent issues in message passing, *oversmoothing* and *oversquashing*. We identify the root cause of these issues as information loss due to *heterophily mixing* in aggregation, where messages of diverse category semantics are mixed. We ...
[]
null
9,699
null
null
ngcZhfXCBW
RLVF: Learning from Verbal Feedback without Overgeneralization
https://openreview.net/forum?id=ngcZhfXCBW
[ "Moritz Pascal Stephan", "Alexander Khazatsky", "Eric Mitchell", "Annie S Chen", "Sheryl Hsu", "Archit Sharma", "Chelsea Finn" ]
Poster
null
The diversity of contexts in which large language models (LLMs) are deployed requires the ability to modify or customize default model behaviors to incorporate nuanced requirements and preferences. A convenient interface to specify such model adjustments is high-level verbal feedback, such as “Don’t use emojis when dra...
[]
null
9,698
2402.10893
title_snapshot
jdRIaUu3xY
BBox-Adapter: Lightweight Adapting for Black-Box Large Language Models
https://openreview.net/forum?id=jdRIaUu3xY
[ "Haotian Sun", "Yuchen Zhuang", "Wei Wei", "Chao Zhang", "Bo Dai" ]
Spotlight
null
Adapting state-of-the-art Large Language Models (LLMs) like GPT-4 and Gemini for specific tasks is challenging. Due to the opacity in their parameters, embeddings, and even output probabilities, existing fine-tuning adaptation methods are inapplicable. Consequently, adapting these black-box LLMs is only possible throug...
[]
null
9,696
2402.08219
title_snapshot
QAGRPiC3FS
RigorLLM: Resilient Guardrails for Large Language Models against Undesired Content
https://openreview.net/forum?id=QAGRPiC3FS
[ "Zhuowen Yuan", "Zidi Xiong", "Yi Zeng", "Ning Yu", "Ruoxi Jia", "Dawn Song", "Bo Li" ]
Poster
null
Recent advancements in Large Language Models (LLMs) have showcased remarkable capabilities across various tasks in different domains. However, the emergence of biases and the potential for generating harmful content in LLMs, particularly under malicious inputs, pose significant challenges. Current mitigation strategies...
[]
null
9,695
2403.13031
title_snapshot
pVyOchWUBa
Position: Understanding LLMs Requires More Than Statistical Generalization
https://openreview.net/forum?id=pVyOchWUBa
[ "Patrik Reizinger", "Szilvia Ujváry", "Anna Mészáros", "Anna Kerekes", "Wieland Brendel", "Ferenc Huszár" ]
Spotlight
null
The last decade has seen blossoming research in deep learning theory attempting to answer, ``Why does deep learning generalize?" A powerful shift in perspective precipitated this progress: the study of overparametrized models in the interpolation regime. In this paper, we argue that another perspective shift is due, si...
[]
null
9,692
2405.01964
title_snapshot
ghYrfdJfjK
PolySketchFormer: Fast Transformers via Sketching Polynomial Kernels
https://openreview.net/forum?id=ghYrfdJfjK
[ "Praneeth Kacham", "Vahab Mirrokni", "Peilin Zhong" ]
Poster
null
The quadratic time and memory complexity inherent to self-attention mechanisms, with respect to sequence length, presents a critical computational bottleneck in the training and deployment of large-scale Transformer-based language models. Recent theoretical results indicate the intractability of sub-quadratic softmax a...
[]
null
9,690
2310.01655
title_snapshot
64I29YeQdt
Quality-Diversity with Limited Resources
https://openreview.net/forum?id=64I29YeQdt
[ "Ren-Jian Wang", "Ke Xue", "Cong Guan", "Chao Qian" ]
Poster
null
Quality-Diversity (QD) algorithms have emerged as a powerful optimization paradigm with the aim of generating a set of high-quality and diverse solutions. To achieve such a challenging goal, QD algorithms require maintaining a large archive and a large population in each iteration, which brings two main issues, sample ...
[]
null
9,689
2406.03731
title_snapshot
cMige5MK1N
Accelerating Heterogeneous Federated Learning with Closed-form Classifiers
https://openreview.net/forum?id=cMige5MK1N
[ "Eros Fanì", "Raffaello Camoriano", "Barbara Caputo", "Marco Ciccone" ]
Poster
null
Federated Learning (FL) methods often struggle in highly statistically heterogeneous settings. Indeed, non-IID data distributions cause client drift and biased local solutions, particularly pronounced in the final classification layer, negatively impacting convergence speed and accuracy. To address this issue, we intro...
[]
null
9,674
2406.01116
title_snapshot
Gp5F6qzwGK
Iterative Regularized Policy Optimization with Imperfect Demonstrations
https://openreview.net/forum?id=Gp5F6qzwGK
[ "Gong Xudong", "Feng Dawei", "Kele Xu", "Yuanzhao Zhai", "Chengkang Yao", "Weijia Wang", "Bo Ding", "Huaimin Wang" ]
Poster
null
Imitation learning heavily relies on the quality of provided demonstrations. In scenarios where demonstrations are imperfect and rare, a prevalent approach for refining policies is through online fine-tuning with reinforcement learning, in which a Kullback–Leibler (KL) regularization is often employed to stabilize the ...
[]
null
9,665
null
null
B1W712hMBi
NExT: Teaching Large Language Models to Reason about Code Execution
https://openreview.net/forum?id=B1W712hMBi
[ "Ansong Ni", "Miltiadis Allamanis", "Arman Cohan", "Yinlin Deng", "Kensen Shi", "Charles Sutton", "Pengcheng Yin" ]
Poster
null
A fundamental skill among human developers is the ability to understand and reason about program execution. As an example, a programmer can mentally simulate code execution in natural language to debug and repair code (aka. rubber duck debugging). However, large language models (LLMs) of code are typically trained on t...
[]
null
9,658
2404.14662
title_snapshot
FYQIgQWH3d
3D Geometric Shape Assembly via Efficient Point Cloud Matching
https://openreview.net/forum?id=FYQIgQWH3d
[ "Nahyuk Lee", "Juhong Min", "Junha Lee", "Seungwook Kim", "Kanghee Lee", "Jaesik Park", "Minsu Cho" ]
Poster
null
Learning to assemble geometric shapes into a larger target structure is a pivotal task in various practical applications. In this work, we tackle this problem by establishing local correspondences between point clouds of part shapes in both coarse- and fine-levels. To this end, we introduce Proxy Match Transform (PMT),...
[]
null
9,656
2407.10542
title_snapshot
vGHOFeUQi8
Simulation-Based Inference with Quantile Regression
https://openreview.net/forum?id=vGHOFeUQi8
[ "He Jia" ]
Poster
null
We present Neural Quantile Estimation (NQE), a novel Simulation-Based Inference (SBI) method based on conditional quantile regression. NQE autoregressively learns individual one dimensional quantiles for each posterior dimension, conditioned on the data and previous posterior dimensions. Posterior samples are obtained ...
[]
null
9,653
2401.02413
title_snapshot
oRLwyayrh1
DRCT: Diffusion Reconstruction Contrastive Training towards Universal Detection of Diffusion Generated Images
https://openreview.net/forum?id=oRLwyayrh1
[ "Baoying Chen", "Jishen Zeng", "Jianquan Yang", "Rui Yang" ]
Spotlight
null
Diffusion models have made significant strides in visual content generation but also raised increasing demands on generated image detection. Existing detection methods have achieved considerable progress, but they usually suffer a significant decline in accuracy when detecting images generated by an unseen diffusion mo...
[]
null
9,651
null
null
37xFIeYgE0
Explaining Probabilistic Models with Distributional Values
https://openreview.net/forum?id=37xFIeYgE0
[ "Luca Franceschi", "Michele Donini", "Cedric Archambeau", "Matthias Seeger" ]
Spotlight
null
A large branch of explainable machine learning is grounded in cooperative game theory. However, research indicates that game-theoretic explanations may mislead or be hard to interpret. We argue that often there is a critical mismatch between what one wishes to explain (e.g. the output of a classifier) and what current ...
[]
null
9,644
2402.09947
title_snapshot
nDps3Q8j2l
Fourier Controller Networks for Real-Time Decision-Making in Embodied Learning
https://openreview.net/forum?id=nDps3Q8j2l
[ "Hengkai Tan", "Songming Liu", "Kai Ma", "Chengyang Ying", "Xingxing Zhang", "Hang Su", "Jun Zhu" ]
Poster
null
Transformer has shown promise in reinforcement learning to model time-varying features for obtaining generalized low-level robot policies on diverse robotics datasets in embodied learning. However, it still suffers from the issues of low data efficiency and high inference latency. In this paper, we propose to investiga...
[]
null
9,642
2405.19885
title_snapshot
2cXzNDe614
PDHG-Unrolled Learning-to-Optimize Method for Large-Scale Linear Programming
https://openreview.net/forum?id=2cXzNDe614
[ "Bingheng Li", "Linxin Yang", "Yupeng Chen", "Senmiao Wang", "Haitao Mao", "Qian Chen", "Yao Ma", "Akang Wang", "Tian Ding", "Jiliang Tang", "Ruoyu Sun" ]
Poster
null
Solving large-scale linear programming (LP) problems is an important task in various areas such as communication networks, power systems, finance and logistics. Recently, two distinct approaches have emerged to expedite LP solving: (i) First-order methods (FOMs); (ii) Learning to optimize (L2O). In this work, we propos...
[]
null
9,641
2406.01908
title_snapshot
5kGfm3Pa41
Recurrent Distance Filtering for Graph Representation Learning
https://openreview.net/forum?id=5kGfm3Pa41
[ "Yuhui Ding", "Antonio Orvieto", "Bobby He", "Thomas Hofmann" ]
Poster
null
Graph neural networks based on iterative one-hop message passing have been shown to struggle in harnessing the information from distant nodes effectively. Conversely, graph transformers allow each node to attend to all other nodes directly, but lack graph inductive bias and have to rely on ad-hoc positional encoding. I...
[]
null
9,640
2312.01538
title_snapshot
9U29U3cDKq
Adaptively Perturbed Mirror Descent for Learning in Games
https://openreview.net/forum?id=9U29U3cDKq
[ "Kenshi Abe", "Kaito Ariu", "Mitsuki Sakamoto", "Atsushi Iwasaki" ]
Poster
null
This paper proposes a payoff perturbation technique for the Mirror Descent (MD) algorithm in games where the gradient of the payoff functions is monotone in the strategy profile space, potentially containing additive noise. The optimistic family of learning algorithms, exemplified by optimistic MD, successfully achieve...
[]
null
9,636
2305.16610
title_snapshot
8ERo4jph0A
Attack-free Evaluating and Enhancing Adversarial Robustness on Categorical Data
https://openreview.net/forum?id=8ERo4jph0A
[ "Yujun Zhou", "Yufei Han", "Haomin Zhuang", "Hongyan Bao", "Xiangliang Zhang" ]
Poster
null
Research on adversarial robustness has predominantly focused on continuous inputs, leaving categorical inputs, especially tabular attributes, less examined. To echo this challenge, our work aims to evaluate and enhance the robustness of classification over categorical attributes against adversarial perturbations throug...
[]
null
9,630
null
null
5kXNMDpUVF
A New Linear Scaling Rule for Private Adaptive Hyperparameter Optimization
https://openreview.net/forum?id=5kXNMDpUVF
[ "Ashwinee Panda", "Xinyu Tang", "Saeed Mahloujifar", "Vikash Sehwag", "Prateek Mittal" ]
Poster
null
An open problem in differentially private deep learning is hyperparameter optimization (HPO). DP-SGD introduces new hyperparameters and complicates existing ones, forcing researchers to painstakingly tune hyperparameters with hundreds of trials, which in turn makes it impossible to account for the privacy cost of HPO w...
[]
null
9,629
2212.04486
title_snapshot
JNN6QHhLHB
Measuring Stochastic Data Complexity with Boltzmann Influence Functions
https://openreview.net/forum?id=JNN6QHhLHB
[ "Nathan Hoyen Ng", "Roger Baker Grosse", "Marzyeh Ghassemi" ]
Poster
null
Estimating the uncertainty of a model’s prediction on a test point is a crucial part of ensuring reliability and calibration under distribution shifts.A minimum description length approach to this problem uses the predictive normalized maximum likelihood (pNML) distribution, which considers every possible label for a d...
[]
null
9,617
2406.02745
title_snapshot
17ZwoHl65h
PlanDQ: Hierarchical Plan Orchestration via D-Conductor and Q-Performer
https://openreview.net/forum?id=17ZwoHl65h
[ "Chang Chen", "Junyeob Baek", "Fei Deng", "Kenji Kawaguchi", "Caglar Gulcehre", "Sungjin Ahn" ]
Poster
null
Despite the recent advancements in offline RL, no unified algorithm could achieve superior performance across a broad range of tasks. Offline *value function learning*, in particular, struggles with sparse-reward, long-horizon tasks due to the difficulty of solving credit assignment and extrapolation errors that accumu...
[]
null
9,611
2406.06793
title_snapshot
ia5XvxFUJT
Gated Linear Attention Transformers with Hardware-Efficient Training
https://openreview.net/forum?id=ia5XvxFUJT
[ "Songlin Yang", "Bailin Wang", "Yikang Shen", "Rameswar Panda", "Yoon Kim" ]
Poster
null
Transformers with linear attention allow for efficient parallel training but can simultaneously be formulated as an RNN with 2D (matrix-valued) hidden states, thus enjoying linear-time inference complexity. However, linear attention generally underperforms ordinary softmax attention. Moreover, current implementations o...
[]
null
9,608
2312.06635
title_snapshot
FQQ4476dT2
FightLadder: A Benchmark for Competitive Multi-Agent Reinforcement Learning
https://openreview.net/forum?id=FQQ4476dT2
[ "Wenzhe Li", "Zihan Ding", "Seth Karten", "Chi Jin" ]
Poster
null
Recent advances in reinforcement learning (RL) heavily rely on a variety of well-designed benchmarks, which provide environmental platforms and consistent criteria to evaluate existing and novel algorithms. Specifically, in multi-agent RL (MARL), a plethora of benchmarks based on cooperative games have spurred the deve...
[]
null
9,606
2406.02081
title_snapshot
Ln3moCobjO
Relaxing the Accurate Imputation Assumption in Doubly Robust Learning for Debiased Collaborative Filtering
https://openreview.net/forum?id=Ln3moCobjO
[ "Haoxuan Li", "Chunyuan Zheng", "Shuyi Wang", "Kunhan Wu", "Hao Wang", "Peng Wu", "Zhi Geng", "Xu Chen", "Xiao-Hua Zhou" ]
Spotlight
null
Recommender system aims to recommend items or information that may interest users based on their behaviors and preferences. However, there may be sampling selection bias in the data collection process, i.e., the collected data is not a representative of the target population. Many debiasing methods are developed based ...
[]
null
9,604
null
null
c0LoolDFw4
A Language Model’s Guide Through Latent Space
https://openreview.net/forum?id=c0LoolDFw4
[ "Dimitri von Rütte", "Sotiris Anagnostidis", "Gregor Bachmann", "Thomas Hofmann" ]
Poster
null
Concept guidance has emerged as a cheap and simple way to control the behavior of language models by probing their hidden representations for concept vectors and using them to perturb activations at inference time. While the focus of previous work has largely been on *truthfulness*, in this paper we extend this framewo...
[]
null
9,601
2402.14433
title_snapshot
o8AaRKbP9K
Can Looped Transformers Learn to Implement Multi-step Gradient Descent for In-context Learning?
https://openreview.net/forum?id=o8AaRKbP9K
[ "Khashayar Gatmiry", "Nikunj Saunshi", "Sashank J. Reddi", "Stefanie Jegelka", "Sanjiv Kumar" ]
Poster
null
Transformers to do reasoning and few-shot learning, without any fine-tuning, is widely conjectured to stem from their ability to implicitly simulate a multi-step algorithms -- such as gradient descent -- with their weights in a single forward pass. Recently, there has been progress in understanding this complex phenome...
[]
null
9,597
2410.08292
title_snapshot
XLlQb24X2o
Test-Time Degradation Adaptation for Open-Set Image Restoration
https://openreview.net/forum?id=XLlQb24X2o
[ "Yuanbiao Gou", "Haiyu Zhao", "Boyun Li", "Xinyan Xiao", "Xi Peng" ]
Spotlight
null
In contrast to close-set scenarios that restore images from a predefined set of degradations, open-set image restoration aims to handle the unknown degradations that were unforeseen during the pretraining phase, which is less-touched as far as we know. This work study this challenging problem and reveal its essence as ...
[]
null
9,594
2312.02197
title_snapshot
5NTTCCO74S
Regression with Multi-Expert Deferral
https://openreview.net/forum?id=5NTTCCO74S
[ "Anqi Mao", "Mehryar Mohri", "Yutao Zhong" ]
Spotlight
null
Learning to defer with multiple experts is a framework where the learner can choose to defer the prediction to several experts. While this problem has received significant attention in classification contexts, it presents unique challenges in regression due to the infinite and continuous nature of the label space. In t...
[]
null
9,593
2403.19494
title_snapshot
L0VoOdjCUb
Learning with 3D rotations, a hitchhiker's guide to SO(3)
https://openreview.net/forum?id=L0VoOdjCUb
[ "Andreas René Geist", "Jonas Frey", "Mikel Zhobro", "Anna Levina", "Georg Martius" ]
Poster
null
Many settings in machine learning require the selection of a rotation representation. However, choosing a suitable representation from the many available options is challenging. This paper acts as a survey and guide through rotation representations. We walk through their properties that harm or benefit deep learning wi...
[]
null
9,590
2404.11735
title_snapshot
GqsRKEhelH
Indirectly Parameterized Concrete Autoencoders
https://openreview.net/forum?id=GqsRKEhelH
[ "Alfred Nilsson", "Klas Wijk", "Sai bharath chandra Gutha", "Erik Englesson", "Alexandra Hotti", "Carlo Saccardi", "Oskar Kviman", "Jens Lagergren", "Ricardo Vinuesa Motilva", "Hossein Azizpour" ]
Poster
null
Feature selection is a crucial task in settings where data is high-dimensional or acquiring the full set of features is costly. Recent developments in neural network-based embedded feature selection show promising results across a wide range of applications. Concrete Autoencoders (CAEs), considered state-of-the-art in ...
[]
null
9,582
2403.00563
title_snapshot
7uwLvFvpis
Vector Quantization Pretraining for EEG Time Series with Random Projection and Phase Alignment
https://openreview.net/forum?id=7uwLvFvpis
[ "Haokun GUI", "Xiucheng Li", "Xinyang Chen" ]
Poster
null
In this paper, we propose a BERT-style self-supervised learning model, VQ-MTM (Vector Quantization Masked Time-Series Modeling), for the EEG time series data analysis. At its core, VQ-MTM comprises a theoretically grounded random-projection quantization module and a phase-aligning module guided by the Time-Phase-Shift ...
[]
null
9,569
null
null
wmljUnbjy6
Unsupervised Parameter-free Simplicial Representation Learning with Scattering Transforms
https://openreview.net/forum?id=wmljUnbjy6
[ "Hiren Madhu", "Sravanthi Gurugubelli", "Sundeep Prabhakar Chepuri" ]
Poster
null
Simplicial neural network models are becoming popular for processing and analyzing higher-order graph data, but they suffer from high training complexity and dependence on task-specific labels. To address these challenges, we propose simplicial scattering networks (SSNs), a parameter-free model inspired by scattering t...
[]
null
9,568
null
null
MWZWUyfFHC
TinyTrain: Resource-Aware Task-Adaptive Sparse Training of DNNs at the Data-Scarce Edge
https://openreview.net/forum?id=MWZWUyfFHC
[ "Young D. Kwon", "Rui Li", "Stylianos Venieris", "Jagmohan Chauhan", "Nicholas Donald Lane", "Cecilia Mascolo" ]
Poster
null
On-device training is essential for user personalisation and privacy. With the pervasiveness of IoT devices and microcontroller units (MCUs), this task becomes more challenging due to the constrained memory and compute resources, and the limited availability of labelled user data. Nonetheless, prior works neglect the d...
[]
null
9,556
2307.09988
title_snapshot
beXQVQorse
High-Dimensional Bayesian Optimization via Semi-Supervised Learning with Optimized Unlabeled Data Sampling
https://openreview.net/forum?id=beXQVQorse
[ "Yuxuan Yin", "Yu Wang", "Peng Li" ]
Spotlight
null
We introduce a novel semi-supervised learning approach, named Teacher-Student Bayesian Optimization ($\texttt{TSBO}$), integrating the teacher-student paradigm into BO to minimize expensive labeled data queries for the first time. $\texttt{TSBO}$ incorporates a teacher model, an unlabeled data sampler, and a student mo...
[]
null
9,555
2305.02614
title_snapshot
NeotatlYOL
SpikeZIP-TF: Conversion is All You Need for Transformer-based SNN
https://openreview.net/forum?id=NeotatlYOL
[ "kang you", "Zekai Xu", "Chen Nie", "Zhijie Deng", "Qinghai Guo", "Xiang Wang", "Zhezhi He" ]
Poster
null
Spiking neural network (SNN) has attracted great attention due to its characteristic of high efficiency and accuracy. Currently, the ANN-to-SNN conversion methods can obtain ANN on-par accuracy SNN with ultra-low latency (8 time-steps) in CNN structure on computer vision (CV) tasks. However, as Transformer-based networ...
[]
null
9,550
2406.03470
title_snapshot
9ZkUFSwlUH
Acquiring Diverse Skills using Curriculum Reinforcement Learning with Mixture of Experts
https://openreview.net/forum?id=9ZkUFSwlUH
[ "Onur Celik", "Aleksandar Taranovic", "Gerhard Neumann" ]
Poster
null
Reinforcement learning (RL) is a powerful approach for acquiring a good-performing policy. However, learning diverse skills is challenging in RL due to the commonly used Gaussian policy parameterization. We propose Diverse Skill Learning (Di-SkilL), an RL method for learning diverse skills using Mixture of Experts, whe...
[]
null
9,547
2403.06966
title_snapshot
Rp8R9C0Sth
AutoOS: Make Your OS More Powerful by Exploiting Large Language Models
https://openreview.net/forum?id=Rp8R9C0Sth
[ "Huilai Chen", "Yuanbo Wen", "Limin Cheng", "Shouxu Kuang", "Yumeng Liu", "Weijia Li", "Ling Li", "Rui Zhang", "Xinkai Song", "Wei Li", "Qi Guo", "Yunji Chen" ]
Poster
null
With the rapid development of Artificial Intelligence of Things (AIoT), customizing and optimizing operating system (OS) kernel configurations for various AIoT application scenarios is crucial for maximizing system performance. However, existing approaches falter due to the overwhelming problem complexity (i.e., over 1...
[]
null
9,546
null
null
efzkSbpyRw
Conformal Predictions under Markovian Data
https://openreview.net/forum?id=efzkSbpyRw
[ "Frédéric Zheng", "Alexandre Proutiere" ]
Poster
null
We study the split Conformal Prediction method when applied to Markovian data. We quantify the gap in terms of coverage induced by the correlations in the data (compared to exchangeable data). This gap strongly depends on the mixing properties of the underlying Markov chain, and we prove that it typically scales as $\s...
[]
null
9,543
2407.15277
title_snapshot
hKdJPMQvew
Hyperbolic Active Learning for Semantic Segmentation under Domain Shift
https://openreview.net/forum?id=hKdJPMQvew
[ "Luca Franco", "Paolo Mandica", "Konstantinos Kallidromitis", "Devin Guillory", "Yu-Teng Li", "Trevor Darrell", "Fabio Galasso" ]
Poster
null
We introduce a hyperbolic neural network approach to pixel-level active learning for semantic segmentation. Analysis of the data statistics leads to a novel interpretation of the hyperbolic radius as an indicator of data scarcity. In HALO (Hyperbolic Active Learning Optimization), for the first time, we propose the use...
[]
null
9,542
2306.11180
title_snapshot
DkqiId4AuR
Failures Are Fated, But Can Be Faded: Characterizing and Mitigating Unwanted Behaviors in Large-Scale Vision and Language Models
https://openreview.net/forum?id=DkqiId4AuR
[ "Som Sagar", "Aditya Taparia", "Ransalu Senanayake" ]
Spotlight
null
In large deep neural networks that seem to perform surprisingly well on many tasks, we also observe a few failures related to accuracy, social biases, and alignment with human values, among others. Therefore, before deploying these models, it is crucial to characterize this failure landscape for engineers to debug and ...
[]
null
9,539
2406.07145
title_snapshot
a9bzTv9SzO
Rolling Diffusion Models
https://openreview.net/forum?id=a9bzTv9SzO
[ "David Ruhe", "Jonathan Heek", "Tim Salimans", "Emiel Hoogeboom" ]
Poster
null
Diffusion models have recently been increasingly applied to temporal data such as video, fluid mechanics simulations, or climate data. These methods generally treat subsequent frames equally regarding the amount of noise in the diffusion process. This paper explores Rolling Diffusion: a new approach that uses a sliding...
[]
null
9,536
2402.09470
title_snapshot
a2uFstsHPb
Efficient Pareto Manifold Learning with Low-Rank Structure
https://openreview.net/forum?id=a2uFstsHPb
[ "Weiyu Chen", "James Kwok" ]
Spotlight
null
Multi-task learning, which optimizes performance across multiple tasks, is inherently a multi-objective optimization problem. Various algorithms are developed to provide discrete trade-off solutions on the Pareto front. Recently, continuous Pareto front approximations using a linear combination of base networks have em...
[]
null
9,531
2407.20734
title_snapshot
OS0szhkPmF
Disentangled Graph Self-supervised Learning for Out-of-Distribution Generalization
https://openreview.net/forum?id=OS0szhkPmF
[ "Haoyang Li", "Xin Wang", "Zeyang Zhang", "Haibo Chen", "Ziwei Zhang", "Wenwu Zhu" ]
Poster
null
Graph out-of-distribution (OOD) generalization, aiming to generalize graph neural networks (GNNs) under distribution shifts between training and testing environments, has attracted ever-increasing attention recently. However, existing literature heavily relies on sufficient task-dependent graph labels, which are often ...
[]
null
9,527
null
null
VyoY3Wh9Wd
In-Context Freeze-Thaw Bayesian Optimization for Hyperparameter Optimization
https://openreview.net/forum?id=VyoY3Wh9Wd
[ "Herilalaina Rakotoarison", "Steven Adriaensen", "Neeratyoy Mallik", "Samir Garibov", "Eddie Bergman", "Frank Hutter" ]
Poster
null
With the increasing computational costs associated with deep learning, automated hyperparameter optimization methods, strongly relying on black-box Bayesian optimization (BO), face limitations. Freeze-thaw BO offers a promising grey-box alternative, strategically allocating scarce resources incrementally to different ...
[]
null
9,512
2404.16795
title_snapshot
st2BTty53v
Transferable Facial Privacy Protection against Blind Face Restoration via Domain-Consistent Adversarial Obfuscation
https://openreview.net/forum?id=st2BTty53v
[ "Kui Zhang", "Hang Zhou", "Jie Zhang", "Wenbo Zhou", "Weiming Zhang", "Nenghai Yu" ]
Poster
null
With the rise of social media and the proliferation of facial recognition surveillance, concerns surrounding privacy have escalated significantly. While numerous studies have concentrated on safeguarding users against unauthorized face recognition, a new and often overlooked issue has emerged due to advances in facial ...
[]
null
9,508
null
null
Bq2THeNXRr
Selecting Large Language Model to Fine-tune via Rectified Scaling Law
https://openreview.net/forum?id=Bq2THeNXRr
[ "Haowei Lin", "Baizhou Huang", "Haotian Ye", "Qinyu Chen", "Zihao Wang", "Sujian Li", "Jianzhu Ma", "Xiaojun Wan", "James Zou", "Yitao Liang" ]
Poster
null
The ever-growing ecosystem of LLMs has posed a challenge in selecting the most appropriate pre-trained model to fine-tune amidst a sea of options. Given constrained resources, fine-tuning all models and making selections afterward is unrealistic. In this work, we formulate this resource-constrained selection task into ...
[]
null
9,501
2402.02314
title_snapshot
YNbCbcGyXE
What Can Transformer Learn with Varying Depth? Case Studies on Sequence Learning Tasks
https://openreview.net/forum?id=YNbCbcGyXE
[ "Xingwu Chen", "Difan Zou" ]
Poster
null
We study the capabilities of the transformer architecture with varying depth. Specifically, we designed a novel set of sequence learning tasks to systematically evaluate and comprehend how the depth of transformer affects its ability to perform memorization, reasoning, generalization, and contextual generalization. We ...
[]
null
9,499
2404.01601
title_snapshot
CgO2cuWWLV
Provable Interactive Learning with Hindsight Instruction Feedback
https://openreview.net/forum?id=CgO2cuWWLV
[ "Dipendra Misra", "Aldo Pacchiano", "Robert E. Schapire" ]
Poster
null
We study interactive learning in a setting where the agent has to generate a response (e.g., an action or trajectory) given a context and an instruction. In contrast, to typical approaches that train the system using reward or expert supervision on response, we study _learning with hindsight labeling_ where a teacher p...
[]
null
9,492
2404.09123
title_snapshot
0ksNeD1SJT
Scaling Exponents Across Parameterizations and Optimizers
https://openreview.net/forum?id=0ksNeD1SJT
[ "Katie E Everett", "Lechao Xiao", "Mitchell Wortsman", "Alexander A Alemi", "Roman Novak", "Peter J Liu", "Izzeddin Gur", "Jascha Sohl-Dickstein", "Leslie Pack Kaelbling", "Jaehoon Lee", "Jeffrey Pennington" ]
Poster
null
Robust and effective scaling of models from small to large width typically requires the precise adjustment of many algorithmic and architectural details, such as parameterization and optimizer choices. In this work, we propose a new perspective on parameterization by investigating a key assumption in prior work about t...
[]
null
9,489
2407.05872
title_snapshot
7tyAO5tUF8
Synergistic Integration of Coordinate Network and Tensorial Feature for Improving Neural Radiance Fields from Sparse Inputs
https://openreview.net/forum?id=7tyAO5tUF8
[ "Mingyu Kim", "Kim Jun-Seong", "Se-Young Yun", "Jin-Hwa Kim" ]
Poster
null
The multi-plane representation has been highlighted for its fast training and inference across static and dynamic neural radiance fields. This approach constructs relevant features via projection onto learnable grids and interpolating adjacent vertices. However, it has limitations in capturing low-frequency details and...
[]
null
9,481
2405.07857
title_snapshot
RbiBKPtuHp
Improving Transformers with Dynamically Composable Multi-Head Attention
https://openreview.net/forum?id=RbiBKPtuHp
[ "Da Xiao", "Qingye Meng", "Shengping Li", "xingyuan yuan" ]
Oral
null
Multi-Head Attention (MHA) is a key component of Transformer. In MHA, attention heads work independently, causing problems such as low-rank bottleneck of attention score matrices and head redundancy. We propose Dynamically Composable Multi-Head Attention (DCMHA), a parameter and computation efficient attention architec...
[]
null
9,473
2405.08553
title_snapshot
t82Y3fmRtk
Training Large Language Models for Reasoning through Reverse Curriculum Reinforcement Learning
https://openreview.net/forum?id=t82Y3fmRtk
[ "Zhiheng Xi", "Wenxiang Chen", "Boyang Hong", "Senjie Jin", "Rui Zheng", "Wei He", "Yiwen Ding", "Shichun Liu", "Xin Guo", "Junzhe Wang", "Honglin Guo", "Wei Shen", "Xiaoran Fan", "Yuhao Zhou", "Shihan Dou", "Xiao Wang", "Xinbo Zhang", "peng sun", "Tao Gui", "Qi Zhang", "Xuan...
Poster
null
In this paper, we propose **R**$^3$: Learning **R**easoning through **R**everse Curriculum **R**einforcement Learning (RL), a novel method that employs only outcome supervision to achieve the benefits of process supervision for large language models. The core challenge in applying RL to complex reasoning is to identify...
[]
null
9,461
2402.05808
title_snapshot
DWT9uiGjxT
Localizing Task Information for Improved Model Merging and Compression
https://openreview.net/forum?id=DWT9uiGjxT
[ "Ke Wang", "Nikolaos Dimitriadis", "Guillermo Ortiz-Jimenez", "François Fleuret", "Pascal Frossard" ]
Poster
null
Model merging and task arithmetic have emerged as promising scalable approaches to merge multiple single-task checkpoints to one multi-task model, but their applicability is reduced by significant performance loss. Previous works have linked these drops to interference in the weight space and erasure of important task-...
[]
null
9,460
2405.07813
title_snapshot
FSxTEvuFa7
CarbonNovo: Joint Design of Protein Structure and Sequence Using a Unified Energy-based Model
https://openreview.net/forum?id=FSxTEvuFa7
[ "Milong Ren", "Tian Zhu", "Haicang Zhang" ]
Poster
null
De novo protein design aims to create novel protein structures and sequences unseen in nature. Recent structure-oriented design methods typically employ a two-stage strategy, where structure design and sequence design modules are trained separately, and the backbone structures and sequences are generated sequentially i...
[]
null
9,454
null
null
M4Htd52HMH
Embodied CoT Distillation From LLM To Off-the-shelf Agents
https://openreview.net/forum?id=M4Htd52HMH
[ "Wonje Choi", "Woo Kyung Kim", "Minjong Yoo", "Honguk Woo" ]
Poster
null
We address the challenge of utilizing large language models (LLMs) for complex embodied tasks, in the environment where decision-making systems operate timely on capacity-limited, off-the-shelf devices. We present DeDer, a framework for decomposing and distilling the embodied reasoning capabilities from LLMs to efficie...
[]
null
9,451
2412.11499
title_snapshot
sF9epWkNUG
Vectorized Conditional Neural Fields: A Framework for Solving Time-dependent Parametric Partial Differential Equations
https://openreview.net/forum?id=sF9epWkNUG
[ "Jan Hagnberger", "Marimuthu Kalimuthu", "Daniel Musekamp", "Mathias Niepert" ]
Poster
null
Transformer models are increasingly used for solving Partial Differential Equations (PDEs). Several adaptations have been proposed, all of which suffer from the typical problems of Transformers, such as quadratic memory and time complexity. Furthermore, all prevalent architectures for PDE solving lack at least one of s...
[]
null
9,450
2406.03919
title_snapshot
v7I5FtL2pV
Tabular Insights, Visual Impacts: Transferring Expertise from Tables to Images
https://openreview.net/forum?id=v7I5FtL2pV
[ "Jun-Peng Jiang", "Han-Jia Ye", "Leye Wang", "Yang Yang", "Yuan Jiang", "De-Chuan Zhan" ]
Spotlight
null
Transferring knowledge across diverse data modalities is receiving increasing attention in machine learning. This paper tackles the task of leveraging expert-derived, yet expensive, tabular data to enhance image-based predictions when tabular data is unavailable during inference. The primary challenges stem from the in...
[]
null
9,447
null
null
p1kDNFs62o
Nesting Particle Filters for Experimental Design in Dynamical Systems
https://openreview.net/forum?id=p1kDNFs62o
[ "Sahel Iqbal", "Adrien Corenflos", "Simo Särkkä", "Hany Abdulsamad" ]
Poster
null
In this paper, we propose a novel approach to Bayesian experimental design for non-exchangeable data that formulates it as risk-sensitive policy optimization. We develop the Inside-Out SMC$^2$ algorithm, a nested sequential Monte Carlo technique to infer optimal designs, and embed it into a particle Markov chain Monte ...
[]
null
9,443
2402.07868
title_snapshot
qGEEso256L
Structure-Aware E(3)-Invariant Molecular Conformer Aggregation Networks
https://openreview.net/forum?id=qGEEso256L
[ "Duy Minh Ho Nguyen", "Nina Lukashina", "Tai Nguyen", "An Thai Le", "TrungTin Nguyen", "Nhat Ho", "Jan Peters", "Daniel Sonntag", "Viktor Zaverkin", "Mathias Niepert" ]
Poster
null
A molecule’s 2D representation consists of its atoms, their attributes, and the molecule’s covalent bonds. A 3D (geometric) representation of a molecule is called a conformer and consists of its atom types and Cartesian coordinates. Every conformer has a potential energy, and the lower this energy, the more likely it o...
[]
null
9,441
2402.01975
title_snapshot
fVg9YrSllr
Beyond ELBOs: A Large-Scale Evaluation of Variational Methods for Sampling
https://openreview.net/forum?id=fVg9YrSllr
[ "Denis Blessing", "Xiaogang Jia", "Johannes Esslinger", "Francisco Vargas", "Gerhard Neumann" ]
Poster
null
Monte Carlo methods, Variational Inference, and their combinations play a pivotal role in sampling from intractable probability distributions. However, current studies lack a unified evaluation framework, relying on disparate performance measures and limited method comparisons across diverse tasks, complicating the ass...
[]
null
9,437
2406.07423
title_snapshot
9yADTDHgGu
When is Transfer Learning Possible?
https://openreview.net/forum?id=9yADTDHgGu
[ "My Phan", "Kianté Brantley", "Stephanie Milani", "Soroush Mehri", "Gokul Swamy", "Geoffrey J. Gordon" ]
Poster
null
We present a general framework for transfer learning that is flexible enough to capture transfer in supervised, reinforcement, and imitation learning. Our framework enables new insights into the fundamental question of *when* we can successfully transfer learned information across problems. We model the learner as inte...
[]
null
9,435
null
null
Lg8nw3ltvX
Layerwise Proximal Replay: A Proximal Point Method for Online Continual Learning
https://openreview.net/forum?id=Lg8nw3ltvX
[ "Jinsoo Yoo", "Yunpeng Liu", "Frank Wood", "Geoff Pleiss" ]
Poster
null
In online continual learning, a neural network incrementally learns from a non-i.i.d. data stream. Nearly all online continual learning methods employ experience replay to simultaneously prevent catastrophic forgetting and underfitting on past data. Our work demonstrates a limitation of this approach: neural networks t...
[]
null
9,433
2402.09542
title_snapshot
CHz7WshPcp
Longitudinal Targeted Minimum Loss-based Estimation with Temporal-Difference Heterogeneous Transformer
https://openreview.net/forum?id=CHz7WshPcp
[ "Toru Shirakawa", "Yi Li", "Yulun Wu", "Sky Qiu", "Yuxuan Li", "Mingduo Zhao", "Hiroyasu Iso", "Mark J. van der Laan" ]
Poster
null
We propose Deep Longitudinal Targeted Minimum Loss-based Estimation (Deep LTMLE), a novel approach to estimate the counterfactual mean of outcome under dynamic treatment policies in longitudinal problem settings. Our approach utilizes a transformer architecture with heterogeneous type embedding trained using temporal-d...
[]
null
9,431
2404.04399
title_snapshot
vYYIuJDTHq
Partial Optimality in the Linear Ordering Problem
https://openreview.net/forum?id=vYYIuJDTHq
[ "David Stein", "Bjoern Andres" ]
Poster
null
The linear ordering problem consists in finding a linear order $<$ on a finite set $A$ so as to minimize the sum of costs associated with pairs of elements $a, b$ for which $a < b$. The problem is NP-hard and APX-hard. We introduce algorithms for solving the problem *partially* by deciding efficiently for some pairs $(...
[]
null
9,430
null
null
sDjszMb2Ir
LASER: Linear Compression in Wireless Distributed Optimization
https://openreview.net/forum?id=sDjszMb2Ir
[ "Ashok Vardhan Makkuva", "Marco Bondaschi", "Thijs Vogels", "Martin Jaggi", "Hyeji Kim", "Michael Gastpar" ]
Poster
null
Data-parallel SGD is the de facto algorithm for distributed optimization, especially for large scale machine learning. Despite its merits, communication bottleneck is one of its persistent issues. Most compression schemes to alleviate this either assume noiseless communication links, or fail to achieve good performance...
[]
null
9,429
2310.13033
title_snapshot
MsnJl6JkZS
Easing Concept Bleeding in Diffusion via Entity Localization and Anchoring
https://openreview.net/forum?id=MsnJl6JkZS
[ "Jiewei Zhang", "Song Guo", "Peiran Dong", "Jie ZHANG", "Ziming Liu", "Yue Yu", "Xiao-Ming Wu" ]
Poster
null
Recent diffusion models have manifested extraordinary capabilities in generating high-quality, diverse, and innovative images guided by textual prompts. Nevertheless, these state-of-the-art models may encounter the challenge of concept bleeding when generating images with multiple entities or attributes in the prompt, ...
[]
null
9,428
null
null
87ZrVHDqmR
Unlocking the Power of Spatial and Temporal Information in Medical Multimodal Pre-training
https://openreview.net/forum?id=87ZrVHDqmR
[ "Jinxia Yang", "Bing Su", "Xin Zhao", "Ji-Rong Wen" ]
Poster
null
Medical vision-language pre-training methods mainly leverage the correspondence between paired medical images and radiological reports. Although multi-view spatial images and temporal sequences of image-report pairs are available in off-the-shelf multi-modal medical datasets, most existing methods have not thoroughly t...
[]
null
9,426
2405.19654
title_snapshot
LWRI4uPG2X
eCeLLM: Generalizing Large Language Models for E-commerce from Large-scale, High-quality Instruction Data
https://openreview.net/forum?id=LWRI4uPG2X
[ "Bo Peng", "Xinyi Ling", "Ziru Chen", "Huan Sun", "Xia Ning" ]
Poster
null
With tremendous efforts on developing effective e-commerce models, conventional e-commerce models show limited success in generalist e-commerce modeling, and suffer from unsatisfactory performance on new users and new products – a typical out-of-domain generalization challenge. Meanwhile, large language models (LLMs) d...
[]
null
9,420
2402.08831
title_snapshot
1xKgDANODx
Retrieval-Augmented Score Distillation for Text-to-3D Generation
https://openreview.net/forum?id=1xKgDANODx
[ "Junyoung Seo", "Susung Hong", "Wooseok Jang", "Inès Hyeonsu Kim", "Min-Seop Kwak", "Doyup Lee", "Seungryong Kim" ]
Poster
null
Text-to-3D generation has achieved significant success by incorporating powerful 2D diffusion models, but insufficient 3D prior knowledge also leads to the inconsistency of 3D geometry. Recently, since large-scale multi-view datasets have been released, fine-tuning the diffusion model on the multi-view datasets becomes...
[]
null
9,419
2402.02972
title_snapshot
elF0QoBSFV
NDOT: Neuronal Dynamics-based Online Training for Spiking Neural Networks
https://openreview.net/forum?id=elF0QoBSFV
[ "Haiyan Jiang", "Giulia De Masi", "Huan Xiong", "Bin Gu" ]
Poster
null
Spiking Neural Networks (SNNs) are attracting great attention for their energy-efficient and fast-inference properties in neuromorphic computing. However, the efficient training of deep SNNs poses challenges in gradient calculation due to the non-differentiability of their binary spike-generating activation functions. ...
[]
null
9,417
null
null
F3Ds71Xgo1
Entropy-Reinforced Planning with Large Language Models for Drug Discovery
https://openreview.net/forum?id=F3Ds71Xgo1
[ "Xuefeng Liu", "Chih-chan Tien", "Peng Ding", "Songhao Jiang", "Rick L. Stevens" ]
Poster
null
The objective of drug discovery is to identify chemical compounds that possess specific pharmaceutical properties toward a binding target. Existing large language models (LLMS) can achieve high token matching scores in terms of likelihood for molecule generation. However, relying solely on LLM decoding often results in...
[]
null
9,415
2406.07025
title_snapshot
vsOF7qDNhl
What is the Long-Run Distribution of Stochastic Gradient Descent? A Large Deviations Analysis
https://openreview.net/forum?id=vsOF7qDNhl
[ "Waïss Azizian", "Franck Iutzeler", "Jerome Malick", "Panayotis Mertikopoulos" ]
Poster
null
In this paper, we examine the long-run distribution of stochastic gradient descent (SGD) in general, non-convex problems. Specifically, we seek to understand which regions of the problem's state space are more likely to be visited by SGD, and by how much. Using an approach based on the theory of large deviations and ra...
[]
null
9,413
2406.09241
title_snapshot
yhpDKSw7yA
Provably Robust DPO: Aligning Language Models with Noisy Feedback
https://openreview.net/forum?id=yhpDKSw7yA
[ "Sayak Ray Chowdhury", "Anush Kini", "Nagarajan Natarajan" ]
Poster
null
Learning from preference-based feedback has recently gained traction as a promising approach to align language models with human interests. While these aligned generative models have demonstrated impressive capabilities across various tasks, their dependence on high-quality human preference data poses a bottleneck in p...
[]
null
9,411
2403.00409
title_snapshot
e0SKaKEEdr
Positive Concave Deep Equilibrium Models
https://openreview.net/forum?id=e0SKaKEEdr
[ "Mateusz Gabor", "Tomasz Piotrowski", "Renato L. G. Cavalcante" ]
Poster
null
Deep equilibrium (DEQ) models are widely recognized as a memory efficient alternative to standard neural networks, achieving state-of-the-art performance in language modeling and computer vision tasks. These models solve a fixed point equation instead of explicitly computing the output, which sets them apart from stand...
[]
null
9,407
2402.04029
title_snapshot
LJ34pX1U5g
Collaborative Heterogeneous Causal Inference Beyond Meta-analysis
https://openreview.net/forum?id=LJ34pX1U5g
[ "Tianyu Guo", "Sai Praneeth Karimireddy", "Michael Jordan" ]
Poster
null
Collaboration between different data centers is often challenged by heterogeneity across sites. To account for the heterogeneity, the state-of-the-art method is to re-weight the covariate distributions in each site to match the distribution of the target population. Nevertheless, this method still relies on the concept...
[]
null
9,403
2404.15746
title_snapshot
Pbey7LqBRl
Neural SPH: Improved Neural Modeling of Lagrangian Fluid Dynamics
https://openreview.net/forum?id=Pbey7LqBRl
[ "Artur Toshev", "Jonas A. Erbesdobler", "Nikolaus A. Adams", "Johannes Brandstetter" ]
Poster
null
Smoothed particle hydrodynamics (SPH) is omnipresent in modern engineering and scientific disciplines. SPH is a class of Lagrangian schemes that discretize fluid dynamics via finite material points that are tracked through the evolving velocity field. Due to the particle-like nature of the simulation, graph neural netw...
[]
null
9,392
2402.06275
title_snapshot
CEfr3h68KU
Purifying Quantization-conditioned Backdoors via Layer-wise Activation Correction with Distribution Approximation
https://openreview.net/forum?id=CEfr3h68KU
[ "Boheng Li", "Yishuo Cai", "Jisong Cai", "Yiming Li", "Han Qiu", "Run Wang", "Tianwei Zhang" ]
Poster
null
Model quantization is a compression technique that converts a full-precision model to a more compact low-precision version for better storage. Despite the great success of quantization, recent studies revealed the feasibility of malicious exploiting model quantization via implanting quantization-conditioned backdoors (...
[]
null
9,389
null
null
ZwrfsrCduj
AD3: Implicit Action is the Key for World Models to Distinguish the Diverse Visual Distractors
https://openreview.net/forum?id=ZwrfsrCduj
[ "Yucen Wang", "Shenghua Wan", "Le Gan", "Shuai Feng", "De-Chuan Zhan" ]
Poster
null
Model-based methods have significantly contributed to distinguishing task-irrelevant distractors for visual control. However, prior research has primarily focused on heterogeneous distractors like noisy background videos, leaving homogeneous distractors that closely resemble controllable agents largely unexplored, whic...
[]
null
9,386
2403.09976
title_snapshot
OnkA4zaEU9
Triadic-OCD: Asynchronous Online Change Detection with Provable Robustness, Optimality, and Convergence
https://openreview.net/forum?id=OnkA4zaEU9
[ "Yancheng Huang", "Kai Yang", "Zelin Zhu", "Leian Chen" ]
Poster
null
The primary goal of online change detection (OCD) is to promptly identify changes in the data stream. OCD problem find a wide variety of applications in diverse areas, e.g., security detection in smart grids and intrusion detection in communication networks. Prior research usually assumes precise knowledge of the syste...
[]
null
9,385
2405.02372
title_snapshot
QhqQJqe0Wq
Score identity Distillation: Exponentially Fast Distillation of Pretrained Diffusion Models for One-Step Generation
https://openreview.net/forum?id=QhqQJqe0Wq
[ "Mingyuan Zhou", "Huangjie Zheng", "Zhendong Wang", "Mingzhang Yin", "Hai Huang" ]
Poster
null
We introduce Score identity Distillation (SiD), an innovative data-free method that distills the generative capabilities of pretrained diffusion models into a single-step generator. SiD not only facilitates an exponentially fast reduction in Fréchet inception distance (FID) during distillation but also approaches or ev...
[]
null
9,382
2404.04057
title_snapshot
geajNKab7g
First-Order Manifold Data Augmentation for Regression Learning
https://openreview.net/forum?id=geajNKab7g
[ "Ilya Kaufman", "Omri Azencot" ]
Poster
null
Data augmentation (DA) methods tailored to specific domains generate synthetic samples by applying transformations that are appropriate for the characteristics of the underlying data domain, such as rotations on images and time warping on time series data. In contrast, *domain-independent* approaches, e.g. *mixup*, are...
[]
null
9,378
2406.10914
title_snapshot
JVhUR8q27o
Towards AutoAI: Optimizing a Machine Learning System with Black-box and Differentiable Components
https://openreview.net/forum?id=JVhUR8q27o
[ "Zhiliang Chen", "Chuan-Sheng Foo", "Bryan Kian Hsiang Low" ]
Poster
null
*Machine learning* (ML) models in the real world typically do not exist in isolation. They are usually part of a complex system (e.g., healthcare systems, self-driving cars) containing multiple ML and *black-box* components. The problem of optimizing such systems, which we refer to as *automated AI* (AutoAI), requires ...
[]
null
9,377
null
null
1YsQI04KaN
Antibody Design Using a Score-based Diffusion Model Guided by Evolutionary, Physical and Geometric Constraints
https://openreview.net/forum?id=1YsQI04KaN
[ "Tian Zhu", "Milong Ren", "Haicang Zhang" ]
Poster
null
Antibodies are central proteins in adaptive immune responses, responsible for protecting against viruses and other pathogens. Rational antibody design has proven effective in the diagnosis and treatment of various diseases like cancers and virus infections. While recent diffusion-based generative models show promise in...
[]
null
9,375
null
null
LO4xhXmFal
DE-COP: Detecting Copyrighted Content in Language Models Training Data
https://openreview.net/forum?id=LO4xhXmFal
[ "André Vicente Duarte", "Xuandong Zhao", "Arlindo L. Oliveira", "Lei Li" ]
Poster
null
*How can we detect if copyrighted content was used in the training process of a language model, considering that the training data is typically undisclosed?* We are motivated by the premise that a language model is likely to identify verbatim excerpts from its training text. We propose DE-COP, a method to determine whe...
[]
null
9,355
2402.09910
title_snapshot
RPMTNGMq0O
Dealing With Unbounded Gradients in Stochastic Saddle-point Optimization
https://openreview.net/forum?id=RPMTNGMq0O
[ "Gergely Neu", "Nneka Okolo" ]
Poster
null
We study the performance of stochastic first-order methods for finding saddle points of convex-concave functions. A notorious challenge faced by such methods is that the gradients can grow arbitrarily large during optimization, which may result in instability and divergence. In this paper, we propose a simple and effec...
[]
null
9,354
2402.13903
title_snapshot
nvHlHfjJPe
$H$-Consistency Guarantees for Regression
https://openreview.net/forum?id=nvHlHfjJPe
[ "Anqi Mao", "Mehryar Mohri", "Yutao Zhong" ]
Poster
null
We present a detailed study of $H$-consistency bounds for regression. We first present new theorems that generalize the tools previously given to establish $H$-consistency bounds. This generalization proves essential for analyzing $H$-consistency bounds specific to regression. Next, we prove a series of novel $H$-consi...
[]
null
9,352
2403.19480
title_snapshot
9zlZuAAb08
Quality Diversity through Human Feedback: Towards Open-Ended Diversity-Driven Optimization
https://openreview.net/forum?id=9zlZuAAb08
[ "Li Ding", "Jenny Zhang", "Jeff Clune", "Lee Spector", "Joel Lehman" ]
Poster
null
Reinforcement Learning from Human Feedback (RLHF) has shown potential in qualitative tasks where easily defined performance measures are lacking. However, there are drawbacks when RLHF is commonly used to optimize for average human preferences, especially in generative tasks that demand diverse model responses. Meanwhi...
[]
null
9,342
2310.12103
title_snapshot
QBj7Uurdwf
Learning Useful Representations of Recurrent Neural Network Weight Matrices
https://openreview.net/forum?id=QBj7Uurdwf
[ "Vincent Herrmann", "Francesco Faccio", "Jürgen Schmidhuber" ]
Oral
null
Recurrent Neural Networks (RNNs) are general-purpose parallel-sequential computers. The program of an RNN is its weight matrix. How to learn useful representations of RNN weights that facilitate RNN analysis as well as downstream tasks? While the _mechanistic approach_ directly looks at some RNN's weights to predict it...
[]
null
9,340
2403.11998
title_snapshot
CaxQ5IbHgF
Convergence of Some Convex Message Passing Algorithms to a Fixed Point
https://openreview.net/forum?id=CaxQ5IbHgF
[ "Vaclav Voracek", "Tomas Werner" ]
Spotlight
null
A popular approach to the MAP inference problem in graphical models is to minimize an upper bound obtained from a dual linear programming or Lagrangian relaxation by (block-)coordinate descent. This is also known as convex/convergent message passing; examples are max-sum diffusion and sequential tree-reweighted message...
[]
null
9,330
2403.07004
title_snapshot
rMV86cAOh6
Stochastic Conditional Diffusion Models for Robust Semantic Image Synthesis
https://openreview.net/forum?id=rMV86cAOh6
[ "Juyeon Ko", "Inho Kong", "Dogyun Park", "Hyunwoo J. Kim" ]
Poster
null
Semantic image synthesis (SIS) is a task to generate realistic images corresponding to semantic maps (labels). However, in real-world applications, SIS often encounters noisy user inputs. To address this, we propose Stochastic Conditional Diffusion Model (SCDM), which is a robust conditional diffusion model that featur...
[]
null
9,329
2402.16506
title_snapshot
ohH3sbUue2
Optimal bounds for $\ell_p$ sensitivity sampling via $\ell_2$ augmentation
https://openreview.net/forum?id=ohH3sbUue2
[ "Alexander Munteanu", "Simon Omlor" ]
Poster
null
Data subsampling is one of the most natural methods to approximate a massively large data set by a small representative proxy. In particular, sensitivity sampling received a lot of attention, which samples points proportional to an individual importance measure called sensitivity. This framework reduces in very general...
[]
null
9,327
2406.00328
title_snapshot
HrzQZXzrN2
Predictive Performance Comparison of Decision Policies Under Confounding
https://openreview.net/forum?id=HrzQZXzrN2
[ "Luke Guerdan", "Amanda Lee Coston", "Ken Holstein", "Steven Wu" ]
Poster
null
Predictive models are often introduced to decision-making tasks under the rationale that they improve performance over an existing decision-making policy. However, it is challenging to compare predictive performance against an existing decision-making policy that is generally under-specified and dependent on unobservab...
[]
null
9,319
2404.00848
title_snapshot
SlRcJvf1yd
Nonsmooth Implicit Differentiation: Deterministic and Stochastic Convergence Rates
https://openreview.net/forum?id=SlRcJvf1yd
[ "Riccardo Grazzi", "Massimiliano Pontil", "Saverio Salzo" ]
Poster
null
We study the problem of efficiently computing the derivative of the fixed-point of a parametric nondifferentiable contraction map. This problem has wide applications in machine learning, including hyperparameter optimization, meta-learning and data poisoning attacks. We analyze two popular approaches: iterative differe...
[]
null
9,317
2403.11687
title_snapshot
9DMMvMTDur
EvIL: Evolution Strategies for Generalisable Imitation Learning
https://openreview.net/forum?id=9DMMvMTDur
[ "Silvia Sapora", "Gokul Swamy", "Chris Lu", "Yee Whye Teh", "Jakob Nicolaus Foerster" ]
Poster
null
Often times in imitation learning (IL), the environment we collect expert demonstrations in and the environment we want to deploy our learned policy in aren't exactly the same (e.g. demonstrations collected in simulation but deployment in the real world). Compared to policy-centric approaches to IL like behavioural clo...
[]
null
9,311
2406.11905
title_snapshot
uku9r6RROl
DRED: Zero-Shot Transfer in Reinforcement Learning via Data-Regularised Environment Design
https://openreview.net/forum?id=uku9r6RROl
[ "Samuel Garcin", "James Doran", "Shangmin Guo", "Christopher G. Lucas", "Stefano V Albrecht" ]
Poster
null
Autonomous agents trained using deep reinforcement learning (RL) often lack the ability to successfully generalise to new environments, even when these environments share characteristics with the ones they have encountered during training. In this work, we investigate how the sampling of individual environment instance...
[]
null
9,306
2402.03479
title_snapshot
6Zl9rv6PDx
Causal Action Influence Aware Counterfactual Data Augmentation
https://openreview.net/forum?id=6Zl9rv6PDx
[ "Núria Armengol Urpí", "Marco Bagatella", "Marin Vlastelica", "Georg Martius" ]
Poster
null
Offline data are both valuable and practical resources for teaching robots complex behaviors. Ideally, learning agents should not be constrained by the scarcity of available demonstrations, but rather generalize beyond the training distribution. However, the complexity of real-world scenarios typically requires huge am...
[]
null
9,304
2405.18917
title_snapshot
aP0H8A1ywk
How Smooth Is Attention?
https://openreview.net/forum?id=aP0H8A1ywk
[ "Valérie Castin", "Pierre Ablin", "Gabriel Peyré" ]
Poster
null
Self-attention and masked self-attention are at the heart of Transformers' outstanding success. Still, our mathematical understanding of attention, in particular of its Lipschitz properties — which are key when it comes to analyzing robustness and expressive power — is incomplete. We provide a detailed study of the Lip...
[]
null
9,302
2312.14820
title_snapshot
4Vqr8SRfyX
Case-Based or Rule-Based: How Do Transformers Do the Math?
https://openreview.net/forum?id=4Vqr8SRfyX
[ "Yi Hu", "Xiaojuan Tang", "Haotong Yang", "Muhan Zhang" ]
Poster
null
Despite the impressive performance in a variety of complex tasks, modern large language models (LLMs) still have trouble dealing with some math problems that are simple and intuitive for humans, such as addition. While we can easily learn basic *rules* of addition and apply them to new problems of any length, LLMs stru...
[]
null
9,299
2402.17709
title_snapshot
biE1uHyG0l
Fundamental Limits of Distributed Covariance Matrix Estimation Under Communication Constraints
https://openreview.net/forum?id=biE1uHyG0l
[ "Mohammad Reza Rahmani", "Mohammad Hossein Yassaee", "Mohammad Ali Maddah-Ali", "Mohammad Reza Aref" ]
Poster
null
Estimating high-dimensional covariance matrices is crucial in various domains. This work considers a scenario where two collaborating agents access disjoint dimensions of $m$ samples from a high--dimensional random vector, and they can only communicate a limited number of bits to a central server, which wants to accura...
[]
null
9,298
2507.16953
title_judge