SkillAgentSearch skills...

Colosseum

A pioneering Python package that creates a bridge between theory and practice in tabular reinforcement learning with an eye on the non-tabular setting.

Install / Use

/learn @MichelangeloConserva/Colosseum
About this skill

Quality Score

0/100

Supported Platforms

Universal

README

<p align="center"> <img src="https://raw.githubusercontent.com/MichelangeloConserva/Colosseum/main/imgs/colosseum_banner.svg" style="width:100%" title="hover text"> </p>

Code style: black Python 3.7

Colosseum is a pioneering Python package that creates a bridge between theory and practice in tabular reinforcement learning with an eye on the non-tabular setting.

If you are new to the package, extensive explanations and tutorials are available at project page.

Core capabilities

  • The computation of three theoretical measures of hardness for any given MDP.
  • Empirical study of the properties of hardness measures.
  • Principled benchmarking for tabular algorithms with rigorous hyperparameters optimization.
  • Non-tabular versions of the tabular benchmark for which tabular hardness measures can be computed.
  • Extensive visualizations for MDPs and analysis tools for the agents' performances.

Example gallery

Hardness analysis

<p> <img src="https://raw.githubusercontent.com/MichelangeloConserva/Colosseum/main/imgs/ha.svg" style="width:25%" title="hover text"> </p>

Agent MDP interaction

<p> <img src="https://raw.githubusercontent.com/MichelangeloConserva/Colosseum/main/imgs/regret.svg" style="width:50%" title="hover text"> </p>

MDP visual representations

<p> <img src="https://raw.githubusercontent.com/MichelangeloConserva/Colosseum/main/imgs/deep_sea.svg" style="width:20%" title="hover text"> <img src="https://raw.githubusercontent.com/MichelangeloConserva/Colosseum/main/imgs/frozen_lake.svg" style="width:20%" title="hover text"> <img src="https://raw.githubusercontent.com/MichelangeloConserva/Colosseum/main/imgs/mge.svg" style="width:20%" title="hover text"> <img src="https://raw.githubusercontent.com/MichelangeloConserva/Colosseum/main/imgs/mgr.svg" style="width:20%" title="hover text"> </p>

Markov chain visual representations

<p> <img src="https://raw.githubusercontent.com/MichelangeloConserva/Colosseum/main/imgs/deep_sea_mc.svg" style="width:20%" title="hover text"> <img src="https://raw.githubusercontent.com/MichelangeloConserva/Colosseum/main/imgs/frozen_lake_mc.svg" style="width:20%" title="hover text"> <img src="https://raw.githubusercontent.com/MichelangeloConserva/Colosseum/main/imgs/mge_mc.svg" style="width:20%" title="hover text"> <img src="https://raw.githubusercontent.com/MichelangeloConserva/Colosseum/main/imgs/mgr_mc.svg" style="width:20%" title="hover text"> </p>

Visitation counts

<p > <img src="https://raw.githubusercontent.com/MichelangeloConserva/Colosseum/main/imgs/deep_sea_vc.svg" style="width:20%" title="hover text"> <img src="https://raw.githubusercontent.com/MichelangeloConserva/Colosseum/main/imgs/frozen_lake_vc.svg" style="width:20%" title="hover text"> <img src="https://raw.githubusercontent.com/MichelangeloConserva/Colosseum/main/imgs/deep_sea_vc2.svg" style="width:20%" title="hover text"> <img src="https://raw.githubusercontent.com/MichelangeloConserva/Colosseum/main/imgs/frozen_lake_vc2.svg" style="width:20%" title="hover text"> </p> <p > <img src="https://raw.githubusercontent.com/MichelangeloConserva/Colosseum/main/imgs/mge_vc.svg" style="width:20%" title="hover text"> <img src="https://raw.githubusercontent.com/MichelangeloConserva/Colosseum/main/imgs/mgr_vc.svg" style="width:20%" title="hover text"> <img src="https://raw.githubusercontent.com/MichelangeloConserva/Colosseum/main/imgs/mge_vc2.svg" style="width:20%" title="hover text"> <img src="https://raw.githubusercontent.com/MichelangeloConserva/Colosseum/main/imgs/mgr_vc2.svg" style="width:20%" title="hover text"> </p>
View on GitHub
GitHub Stars13
CategoryEducation
Updated2mo ago
Forks3

Languages

Python

Security Score

90/100

Audited on Jan 10, 2026

No findings