robust AI

2 items

RESEARCHarXiv CS.LG·4/17/2026

Optimistic Policy Learning under Pessimistic Adversaries with Regret and Violation Guarantees

This research addresses the challenge of decision-making in environments with strategic adversaries or external factors, where traditional policies can fail catastrophically in safety-critical settings. It proposes an optimistic policy learning approach designed to account for these interactions and provide regret and violation guarantees.

reinforcement learning robust AI adversarial AI

RESEARCHarXiv CS.LG·4/17/2026

The Devil Is in Gradient Entanglement: Energy-Aware Gradient Coordinator for Robust Generalized Category Discovery

This research paper introduces an Energy-Aware Gradient Coordinator to address "gradient entanglement," a key challenge in Robust Generalized Category Discovery. The proposed method aims to improve the robustness and performance of AI models in identifying new categories.

Gradient Descent category discovery deep learning machine learning