MLE for contextual bandit

The code in this repository shows an example of maximum likelihood estimation (MLE) for choice data in a contextual bandit task.

Example paper with this task and analysis approach:

• Leong YC, Radulescu A, Daniel R, DeWoskin V, Niv Y. Dynamic Interaction between Reinforcement Learning and Attention in Multidimensional Environments. Neuron. 2017 Jan 18;93(2):451-463. doi: 10.1016/j.neuron.2016.12.040. PMID: 28103483; PMCID: PMC5287409.

├── contextual-bandit
│   ├── data
│   ├── fits
│   ├── libraries
│   │   ├── Data.py
│   │   ├── FeatureRL.py
│   │   ├── fitting.py
│   │   ├── World.py
├── ReadMe.md
├── fit-frl-choice.ipynb
└── .gitignore