Category: In-Context RL

• What to do with poor pre-training data? Bengio team introduces explicit Bayesian for gradient-free In-Context RL