v-i-s-h/mab.jl

A Julia Package for providing Multi Armed Bandit Experiments

MAB.jl - A Package for Bandit Experiments

This package provide a framework for developing and comparing various Bandit algorithms

Available Algorithms

0. Uniform Strategy (Randomly picking some arm)
1. ϵ-greedy
   1. ϵ-greedy
   2. ϵ_n greedy
2. Upper Confidence Bound Policies
   1. UCB1
   2. UCB-Normal
   3. UCB-V
   4. Bayes-UCB (For Bernoulli Rewards)
   5. KL-UCB
   6. Discounted-UCB
   7. Sliding Window UCB
3. Thompson Sampling
   1. Thompson Sampling
   2. Dynamic Thompson Sampling
   3. Optimistic Thompson Sampling
   4. TSNormal (Thompson Sampling for Gaussian distributed rewards)
   5. Restarting Thompson Sampling
   6. TS With Gaussian Prior
4. EXP3
   1. EXP3
   2. EXP3.1
   3. EXP3-IX
5. SoftMax
6. REXP3
7. Gradient Bandit

Available Arm Models

1. Bernoulli
2. Beta
3. Normal
4. Sinusoidal (without noise)
5. Pulse (without noise)
6. Square
7. Variational (without noise)

mab.jl's Issues

Example for contextual bandit with thompson sampling

Anyone have an example/tutorial of a contextual bandit using Thompson sampling as mentioned in the comments in source files?

Info about upcoming removal of packages in the General registry

As described in https://discourse.julialang.org/t/ann-plans-for-removing-packages-that-do-not-yet-support-1-0-from-the-general-registry/ we are planning on removing packages that do not support 1.0 from the General registry. This package has been detected to not support 1.0 and is thus slated to be removed. The removal of packages from the registry will happen approximately a month after this issue is open.

To transition to the new Pkg system using Project.toml, see https://github.com/JuliaRegistries/Registrator.jl#transitioning-from-require-to-projecttoml.
To then tag a new version of the package, see https://github.com/JuliaRegistries/Registrator.jl#via-the-github-app.

If you believe this package has erroneously been detected as not supporting 1.0 or have any other questions, don't hesitate to discuss it here or in the thread linked at the top of this post.