Florent Altché

EXPLORATION BY BOOTSTRAPPED PREDICTION

COMPUTER CODE GENERATION FROM TASK DESCRIPTIONS USING NEURAL NETWORKS

REINFORCEMENT LEARNING USING HINDSIGHT TO MODEL UNPREDICTABLE ASPECTS OF THE FUTURE

DISCRETE TOKEN PROCESSING USING DIFFUSION MODELS

LEARNING ENVIRONMENT REPRESENTATIONS FOR AGENT CONTROL USING PREDICTIONS OF BOOTSTRAPPED LATENTS

SELF-SUPERVISED REPRESENTATION LEARNING USING BOOTSTRAPPED LATENT REPRESENTATIONS