# alessiodm/drl-zh

Deep Reinforcement Learning: Zero to Hero!

Repository: https://github.com/alessiodm/drl-zh
Canonical: https://ross.abutalabs.com/products/drl-zh
Language: Jupyter Notebook
License: MIT
License Family: permissive
Topics: deep-reinforcement-learning, reinforcement-learning, deep-learning, machine-learning
Last push: 2026-05-26T16:54:31+00:00

## Health v2 (maintenance only)
Score: 73/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 84, release rhythm 62, longevity 69
- inputs: {"age_days": 966, "days_push": 99, "days_rel": 99, "gap_med": 135.5, "n_releases_24m": 3}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2293, forks 115 (observed 2026-08-28T04:06:35.054570+00:00)

## What it is
A hands-on deep reinforcement learning course delivered as Jupyter notebooks, taking learners from MDPs and tabular RL to modern methods like RLHF, AlphaZero-style planning, world models, and meta-learning. Exercise notebooks contain guided TODOs with full solutions provided, plus a Docker workspace with an AI companion VS Code extension.

## Use cases
- learn deep reinforcement learning from scratch
- implement DQN, PPO, and SAC myself instead of reading about them
- understand how RLHF and DPO train language models
- study AlphaZero-style planning and Monte Carlo Tree Search
- learn world models like Dreamer with worked code
- practice RL with guided exercises and solutions
- get hands-on with multi-agent and offline RL

## When to choose
- you want to build RL algorithms from first principles with guided exercises
- you want a structured curriculum covering both classic and cutting-edge RL including RLHF and world models
- you prefer runnable notebooks with solutions when stuck

## When to avoid
- you need a production-ready RL library rather than educational code
- you want a video course or textbook instead of notebooks
- you need GPU-cluster-scale training out of the box

## Facets
- artifact type: learning-resource
- maturity: active
- function: machine-learning, reinforcement-learning, deep-learning
- domain: reinforcement-learning, machine-learning, deep-learning, tutorials
- platform: python, cross-platform
- tags: jupyter-notebooks, course, rlhf, world-models, alphazero, hands-on-learning, exercises, docker

## Member repositories
- alessiodm/drl-zh (main) score 73

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:35.054570+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:40:27.348387+00:00, confidence not recorded.
  - readme: https://github.com/alessiodm/drl-zh (fetched 2026-08-28T04:06:35.054570+00:00, sha a113b27b960e)
- Data as of 2026-08-30T08:39:29.467469+00:00.