paper
Adversarial Attacks on Neural Network Policies
paperactive
adversarial-attacks-on-neural-network-policies-7f757f72·1 events·first seen 28d agoAliases: Adversarial Attacks on Neural Network Policies
Co-occurring entities
More like this (12)
adversarial examplesblack-box adversarial attacksGiving AI a Headache: Acoustic Adversarial Attacks to Computer Vision Applicationsadversarial trainingadversarial robustnessBeyond Runtime Enforcement: Shield Synthesis as Defensibility Analysis for Adversarial NetworksGenerative Adversarial NetworksLearning Red Agent Policy from Observations for Neurosymbolic Autonomous Cyber AgentsYour Privacy My Cloak: Backdoor Attacks on Differentially Private Federated LearningLatent Adversarial Robustification (LAR)Adversarial Creation and Detection of AI-Generated Social Bot ContentThe Malicious Use of Artificial Intelligence
Recent events (1)
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.