Project for the 2022/2023 Reinforcement Learning exam.
We implemented the curiosity-driven approach from the Curiosity-driven Exploration by Self-supervised Prediction paper.
no_curiosity.mp4
curiosity_level2.mp4
no_curiosity_level2-2.mp4
- [1] Curiosity-driven Exploration by Self-supervised Prediction
- [2] EFFICIENT PARALLEL METHODS FOR DEEP REINFORCEMENT LEARNING
- [3] CCLF: A Contrastive-Curiosity-Driven Learning Framework for Sample-Efficient Reinforcement Learning
- [4] SPATIAL GRAPH ATTENTION AND CURIOSITY-DRIVEN POLICY FOR ANTIVIRAL DRUG DISCOVERY
- [5] ATTENTION-BASED CURIOSITY-DRIVEN EXPLORATION IN DEEP REINFORCEMENT LEARNING
- [6] Proximal Policy Optimization Algorithms
Luigi Antonelli
Florin Cuconasu