Exploration-exploitation is a powerful and practical tool in multi-agent learning (MAL);
however, its effects are far from understood. To make progress in this direction, we study a …

Limitations in acquiring training data restrict potential applications of deep reinforcement
learning (RL) methods to the training of real-world robots. Here we train an RL agent to align …

The current state-of-the-art deep reinforcement learning techniques require agents to gather
large amounts of diverse experiences to train effective and general models. In addition …

Reinforcement learning (RL) algorithms have achieved notable success in recent years, but
still struggle with fundamental issues in long-term credit assignment. It remains difficult to …

Seq2seq models offer great promise for sequence generation problems such as machine
translation, text summarization, or dialogue generation. Nevertheless, fully supervised …