Alexandre Megretski

deep reinforcement learning (1) contrastive learning (1) model misspecification (1) uncertainty quantification (1) epistemic uncertainty (1) policy learning (1) robust reinforcement learning (1) variance estimation (1) adversarial attack (1) latent representation (1) dynamical system (1) nonlinear dynamical system (1) adversarial perturbation (1) online control (1) robust control (1) aleatoric uncertainty (1) deep q-learning (1) physics constraint (1) conservation law (1) neural network (1)