Yann Dubois

representation learning (4) self-supervised learning (3) data augmentation (2) linear probing (2) information bottleneck (1) preference learning (1) reinforcement learning (1) generalization gap (1) natural language processing (1) supervised learning (1) instruction following (1) reinforcement learning from human feedback (1) maximum likelihood (1) human feedback (1) recurrent neural network (1) convolutional neural network (1) generalization bound (1) reward model (1) attention mechanism (1) variational inference (1)