Independent projects — infrastructure, robotics, technology surveys, and corporate research
← Learning guides

Reinforcement Learning Primer: reading guide

MDPs and Bellman equations through deep RL, imitation learning, and Sim-to-Real, always tied to robot applications.

This is an editorially selected reading order. Begin at step 1 or go directly to the topic you need.

  1. Fundamentals · 20 min

    Reinforcement Learning Basics — MDPs, Bellman Equations, and Exploration for Robots

    Reinforcement learning is a closed loop in which an agent chooses actions from observations and maximizes delayed rewards. This primer explains MDPs, value functions, policies, Bellman equations, exploration versus exploitation, reward design, and the path from simulation to a real robot.

  2. Fundamentals · 18 min

    Q-Learning and DQN — From a Q-Table to Deep Reinforcement Learning

    Q-learning updates action values with the Bellman optimality target. This primer follows the path from a tabular Q-table to a Deep Q-Network, explaining experience replay, target networks, overestimation, and safe placement in a robot stack.

  3. Fundamentals · 19 min

    Policy Gradients, PPO, and SAC — Stable Continuous Control for Robots

    Policy gradients update a policy directly so that steering, thrust, and joint torque can remain continuous. This article connects Actor-Critic, PPO, and SAC, then covers clipping, entropy regularization, evaluation, and the safety boundary for Sim-to-Real transfer.

  4. Fundamentals · 27 min

    Reward Design Primer — Why "What to Maximize" Is the Hardest Part of RL

    In reinforcement learning implementations, reward-function design more often determines the outcome than the algorithm does. This article organizes sparse vs. dense reward, the policy invariance of potential-based reward shaping, real cases of reward hacking, inverse reinforcement learning, and constrained RL.

  5. Fundamentals · 13 min

    Model-Based Reinforcement Learning and Sim-to-Real

    World models, prediction error, MPC, domain randomization, system identification, and safety monitoring for real machines.

  6. Fundamentals · 19 min

    Reinforcement Learning Primer: Imitation Learning and Inverse Reinforcement Learning

    Behavior Cloning, DAgger, and inverse reinforcement learning use demonstrations when rewards are hard to write. This primer covers covariate shift, reward inference, VLA connections, evaluation, safety, and data provenance.

  7. Fundamentals · 37 min

    π0 Explained — How Flow Matching Changed VLA Action Generation

    A source-based technical guide to Physical Intelligence's π0: the PaliGemma VLM and dedicated Action Expert, continuous action generation with conditional flow matching, cross-embodiment training across seven robot types, and the difference from autoregressive VLA models such as RT-2 and OpenVLA.

  8. Fundamentals · 31 min

    Multi-Agent Reinforcement Learning Primer — Optimizing in a World Where the Other Side Is Learning Too

    When multiple agents learn at the same time, a single-agent MDP no longer holds. This article organizes non-stationarity, cooperative/competitive/mixed settings, CTDE, the credit-assignment problem, and MADDPG and QMIX, with equations and diagrams.

Search this field →