Almanac
paper

Adversarial Attacks on Neural Network Policies

paperactiveadversarial-attacks-on-neural-network-policies-7f757f72·1 events·first seen 28d ago

Aliases: Adversarial Attacks on Neural Network Policies

Co-occurring entities

More like this (12)

Recent events (1)

4Openai Blog·28d ago·source ↗

Adversarial Attacks on Neural Network Policies

OpenAI published research examining adversarial attacks on neural network-based reinforcement learning policies. The work investigates how small, carefully crafted perturbations to observations can cause trained RL agents to fail catastrophically. This represents an early investigation into the robustness and safety of learned policies under adversarial conditions.