RL Algorithms Cheat Sheet
~~~~~~~~~~~~~~~~~~~~~~~~~~

In this page you can find the algorithms that are currently implemented in DI-engine.
All pages relative to our algorithms are grouped by category and follow the structure below:

- Overview
- Quick facts
- Key Equations or Key Graphs
- Pseudo-code
- Extensions (algorithm improvements and variants)
- Implementations
- Benchmark
- References

.. toctree::
    :maxdepth: 2
    :caption: Q-learning

    dqn
    c51
    qrdqn
    rainbow
    iqn
    fqf
    sql
    sqn
    mdqn
    averaged_dqn

.. toctree::
    :maxdepth: 2
    :caption: Actor-Critic

    a2c
    ppo
    acer
    impala
    ppg
    ddpg
    d4pg
    td3
    sac

.. toctree::
    :maxdepth: 2
    :caption: Exploration

    rnd
    her
    icm

.. toctree::
    :maxdepth: 2
    :caption: Imitation Learning

    dqfd
    sqil
    gail
    trex
    r2d3

.. toctree::
    :maxdepth: 2
    :caption: Offline RL

    cql
    td3_bc
    edac
    dt
    qgpo
    diffuser

.. toctree::
    :maxdepth: 2
    :caption: Memory Based

    r2d2
    gtrxl

.. toctree::
    :maxdepth: 2
    :caption: Multi-Agent

    qmix
    coma
    wqmix
    qtran
    collaq
    atoc

.. toctree::
    :maxdepth: 2
    :caption: Model-Based RL

    mbpo
    vpn

.. toctree::
    :maxdepth: 2
    :caption: Generalization

    plr