A curated list of resources for Learning with Noisy Labels

A curated (most recent) list of resources for Learning with Noisy Labels

Official Implementation of Early-Learning Regularization Prevents Memorization of Noisy Labels

[NeurIPS 2021] WRENCH: Weak supeRvision bENCHmark

[arXiv:2411.10023] "Model Inversion Attacks: A Survey of Approaches and Countermeasures"

[ICML2020] Normalized Loss Functions for Deep Learning with Noisy Labels

Defending graph neural networks against adversarial attacks (NeurIPS 2020)

Code for "Transfer Learning without Knowing: Reprogramming Black-box Machine Learning Models with Scarce Data and Limited Resources". (ICML 2020)

AAAI 2021: Beyond Class-Conditional Assumption: A Primary Attempt to Combat Instance-Dependent Label Noise

A curated list of Robust Machine Learning papers/articles and recent advancements.

AAAI 2021: Robustness of Accuracy Metric and its Inspirations in Learning with Noisy Labels

No description provided.

Principled learning method for Wasserstein distributionally robust optimization with local perturbations (ICML 2020)

Source code for Self-Guided Learning to Denoise for Robust Recommendation. SIGIR 2022.

[Re] Can gradient clipping mitigate label noise? (ML Reproducibility Challenge 2020)

"RDA: Reciprocal Distribution Alignment for Robust Semi-supervised Learning" by Yue Duan (ECCV 2022)

Code for "Adversarial Robustness via Runtime Masking and Cleansing" (ICML 2020)

Adversarial Attacks and Defenses via Image perturbations

[ICML 24] S-DQN and S-PPO: Robust smoothed deep RL agents without sacrificing performance

Corruption Robust Image Classification with a new Activation Function. Our proposed Activation Function is inspired by the Human Visual System and a classic signal processing fix for data corruption.

Mixtures-of-ExperTs modEling for cOmplex and non-noRmal dIsTributionS

Xinshao Wang, Ex-Postdoc and Ex-Visit Scholar@University of Oxford, Ex-Senior Researcher@ZenithAI

State-level optimization framework that combines deep learning with dual-timescale control to adapt latent manifolds and yield robust, self-governing agents in non-stationary environments.

implementation of Claru activation function

Epistemic Weight Engine (EWE) — A pre-update gating mechanism for signal-reliability-weighted learning in AI systems. ACM TIST 2026.