Background Literature

To familiarize yourself with the core concepts, please read the writeup [Ago26].

Deep Reinforcement Learning

Relevant deep reinforcement learning [SB18] methods include:

• Approximate value iteration [AMSB19, Bel57, BT96]

• Deep Q-learning [MKS+15, WD92]

• Hindsight experience replay [AWR+17]

• Limited-horizon Bellman-based learning (LHBL) [HAS26]

Heuristic Search

Relevant heuristic search methods include:

• A* Search: [HNR68]

• Batch weighted A* search: [AMSB19, LCM+22]

• Batch weighted Q* search: [ASS+24]

Supervised Learning

Relevant supervised learning methods include:

• Heuristic supervision from random walks: [CKB+25]

• Policy supervision from random walks: [Tak21]