Aerial View Localization with Reinforcement Learning: Towards Emulating Search-and-Rescue
Comments Accepted to ICLR 2023 Workshop on Machine Learning for Remote Sensing
期刊&会议
ICLR
International Conference on Learning Representations · 会议 · Machine Learning
Indiscriminate Poisoning Attacks on Unsupervised Contrastive Learning
Comments ICLR 2023 Spotlight (notable top 25%). The first two authors contributed equally to this paper
Learn2Agree: Fitting with Multiple Annotators without Objective Ground Truth
Comments Accepted by the TML4H workshop at ICLR 2023
FastFill: Efficient Compatible Model Update
Comments To appear in The Eleventh International Conference on Learning Representations
A Message Passing Perspective on Learning Dynamics of Contrastive Learning
Comments ICLR 2023
Optimal ANN-SNN Conversion for High-accuracy and Ultra-low-latency Spiking Neural Networks
Journal ref International Conference on Learning Representations (2022)
Bort: Towards Explainable Neural Networks with Bounded Orthogonal Constraint
Comments ICLR 2023 accepted
Out-of-Distribution Detection and Selective Generation for Conditional Language Models
Comments Published in ICLR 2023
Self-supervised learning with rotation-invariant kernels
Journal ref The Eleventh International Conference on Learning Representations, May 2023, Kigali, Rwanda
Learning Hyper Label Model for Programmatic Weak Supervision
Comments Published as a conference paper at ICLR 2023
PASHA: Efficient HPO and NAS with Progressive Resource Allocation
Comments Accepted at ICLR 2023
Causal Representation Learning for Instantaneous and Temporal Effects in Interactive Systems
Comments Published at International Conference on Learning Representations (ICLR), 2023
Variance Reduction is an Antidote to Byzantines: Better Rates, Weaker Assumptions and Communication Compression as a Cherry on the Top
Comments ICLR 2023. 42 pages, 8 figures. Changes in v2: few typos and inaccuracies were fixed, more clarifications were added. Changes in v3: ICLR formatting was applied, additional experiments were added (Appendix B.4-B.5) and extra discussion of the results was added to Appendix E.5. Code: https://github.com/SamuelHorvath/VR_Byzantine
RLx2: Training a Sparse Deep Reinforcement Learning Model from Scratch
Comments ICLR 2023 spotlight
Diminishing Return of Value Expansion Methods in Model-Based Reinforcement Learning
Comments Published as a conference paper at ICLR 2023
Proactive Multi-Camera Collaboration For 3D Human Pose Estimation
Comments ICLR 2023 poster
MAST: Masked Augmentation Subspace Training for Generalizable Self-Supervised Priors
Comments ICLR 2023
Neural Compositional Rule Learning for Knowledge Graph Reasoning
Comments ICLR 2023
3D Equivariant Diffusion for Target-Aware Molecule Generation and Affinity Prediction
Comments Accepted to ICLR 2023
Decision Transformer under Random Frame Dropping
Comments 20 pages, 17 figures; accepted to ICLR 2023
Reparameterization through Spatial Gradient Scaling
Comments Published at ICLR 2023. Code available at https://github.com/Ascend-Research/Reparameterization
Grounding Graph Network Simulators using Physical Sensor Observations
Comments Accepted as a poster at the 11th International Conference on Learning Representations (ICLR), 2023
Can discrete information extraction prompts generalize across language models?
Comments Published as conference paper at ICLR 2023
Data Valuation Without Training of a Model
Comments ICLR 2023
Latent Variable Representation for Reinforcement Learning
Comments ICLR 2023. The first two authors contribute equally. Project Website: https://rlrep.github.io/lvrep/
Iterative Patch Selection for High-Resolution Image Recognition
Comments Published as a conference paper at ICLR 2023
Denoising Masked AutoEncoders Help Robust Classification
Comments ICLR 2023
VIP: Towards Universal Visual Reward and Representation via Value-Implicit Pre-Training
Comments ICLR 2023, Notable-Top-25% (Spotlight). Project website: https://sites.google.com/view/vip-rl
Relative representations enable zero-shot latent space communication
Comments ICLR 2023 notable top 5%, 26 pages, 11 figures, 18 tables
Bridging the Gap to Real-World Object-Centric Learning
Comments ICLR 2023 camera-ready version
