POMGs--in which sample-efficient learning is tractable.

Similarly, Multiagent RL (MARL) can also be accelerated if agents can share knowledge with each other.

V100 GPUs to train the models. Consortium and are licensed under a Creative Commons Attribution 4.0 License. Similarly, for evaluating the agent listener with a human speaker, each agent evaluates 400 human utterances in Fig 5b. In Fig 10, we present the results of the human evaluation on the text game. Sec 4.3, we show that agents trained using our method beat all prior baselines when paired with both The blue bars show the standard deviation across all agents present in the buffer.

We address the question of repeatedly learning linear classifiers against agents who are strategically trying to game the deployed classifiers, and we use the Stackelberg regret to measure the performance of our algorithms. First, we show that Stackelberg and external regret for the problem of strategic classification are strongly incompatible: i.e., there exist worst-case scenarios, where any sequence of actions providing sublinear external regret might result in linear Stackelberg regret and vice versa. Second, we present a strategy-aware algorithm for minimizing the Stackelberg regret for which we prove nearly matching upper and lower regret bounds. Finally, we provide simulations to complement our theoretical analysis. Our results advance the growing literature of learning from revealed preferences, which has so far focused on "smoother" assumptions from the perspective of the learner and the agents respectively.