Papers

12094 papers

ICLR2024

On Bias-Variance Alignment in Deep Models

Summary pending...

bias-variance decompositionensembledeep learning
ICLR2024

Learning Hierarchical Polynomials with Three-Layer Neural Networks

Summary pending...

Hierarchical polynomialsfeature learningthree-layer networks
ICLR2024

Negatively Correlated Ensemble Reinforcement Learning for Online Diverse Game Level Generation

Summary pending...

Level GenerationVideo GamesDeep Reinforcement Learning
ICLR2024

Adaptive Regret for Bandits Made Possible: Two Queries Suffice

Summary pending...

adaptive regretmulti arm bandit
ICLR2024

Merge, Then Compress: Demystify Efficient SMoE with Hints from Its Routing Policy

Summary pending...

Sparse Mixture-of-ExpertsEfficiencyMerging
ICLR2024

Making RL with Preference-based Feedback Efficient via Randomization

Summary pending...

reinforcement learningpreference-based feedbacktheory
ICLR2024

Beyond Weisfeiler-Lehman: A Quantitative Framework for GNN Expressiveness

Summary pending...

Graph Neural NetworksExpressive PowerHomomorphism
ICLR2024

Generating Pragmatic Examples to Train Neural Program Synthesizers

Summary pending...

program synthesispragmaticsself-play
ICLR2024

Generative Modeling of Regular and Irregular Time Series Data via Koopman VAEs

Summary pending...

Time Series GenerationKoopman Theory; Variational Autoencoder; Generative Modeling
ICLR2024

Beyond Worst-case Attacks: Robust RL with Adaptive Defense via Non-dominated Policies

Summary pending...

robust reinforcement learning; beyond worse-case
ICLR2024

Lipsum-FT: Robust Fine-Tuning of Zero-Shot Models Using Random Text Guidance

Summary pending...

computer visionvision-langauge modeltransfer learning
ICLR2024

MUFFIN: Curating Multi-Faceted Instructions for Improving Instruction Following

Summary pending...

Instruction TuningLarge Language ModelsAutomatic Data Generation
ICLR2024

Concept Bottleneck Generative Models

Summary pending...

Interpretabilitygenerative models
ICLR2024

Robustifying and Boosting Training-Free Neural Architecture Search

Summary pending...

Neural Architecture SearchTraining-free NASBayesian Optimization
ICLR2024

Score Regularized Policy Optimization through Diffusion Behavior

Summary pending...

offline reinforcement learninggenerative modelsdiffusion models
ICLR2024

Goodhart's Law in Reinforcement Learning

Summary pending...

reinforcement learninggoodhart's lawmisspecification
ICLR2024

Hyper Evidential Deep Learning to Quantify Composite Classification Uncertainty

Summary pending...

Evidential Neural Networkhyperdomainvagueness
ICLR2024

Understanding Domain Generalization: A Noise Robustness Perspective

Summary pending...

out-of-distribution generalizationdistribution shiftsspurious correlation
ICLR2024

DataInf: Efficiently Estimating Data Influence in LoRA-tuned LLMs and Diffusion Models

Summary pending...

Influence functionData valuation
ICLR2024

Flag Aggregator: Scalable Distributed Training under Failures and Augmented Losses using Convex Optimization

Summary pending...

RobustAggregationDistributed