Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / SMPyBandits/SMPyBandits issues and pull requests

#247 - Question on using SMPyBandits on Neuroscience Data

Issue - State: closed - Opened by fspinar 5 months ago - 2 comments

#246 - Update links in README.md

Pull Request - State: closed - Opened by zverianskii 5 months ago - 1 comment

#245 - Delayed Feedback Support for Simulation Configuration

Issue - State: open - Opened by jerrylam almost 2 years ago - 2 comments

#244 - evaluation.plotBestArmPulls(envId) Usage ?

Issue - State: open - Opened by Dexter123193 over 2 years ago

#242 - Minor math error in "Easily_creating_MAB_problems" notebook

Issue - State: closed - Opened by kortukov almost 3 years ago - 1 comment

#241 - Optimized store function by removing repetitions

Pull Request - State: closed - Opened by SlyJabiru almost 3 years ago

#240 - AdSwitchNew: Speed improvement on check_changes_good_arms

Pull Request - State: closed - Opened by balbok0 over 3 years ago - 2 comments

#239 - again

Pull Request - State: closed - Opened by JulSez about 4 years ago

#238 - remove EFFRAWUCB++2

Pull Request - State: closed - Opened by JulSez about 4 years ago

#237 - Asymptotic experiment

Pull Request - State: closed - Opened by JulSez about 4 years ago

#236 - plot

Pull Request - State: closed - Opened by JulSez about 4 years ago

#235 - bug

Pull Request - State: closed - Opened by JulSez about 4 years ago

#234 - plot : last change in plot hopefully

Pull Request - State: closed - Opened by JulSez about 4 years ago

#233 - add RAWUCB++2 to experiment

Pull Request - State: closed - Opened by JulSez about 4 years ago

#232 - add RAWUCB++2

Pull Request - State: closed - Opened by JulSez about 4 years ago

#231 - Rotting/fix

Pull Request - State: closed - Opened by JulSez about 4 years ago

#230 - fix4

Pull Request - State: closed - Opened by JulSez about 4 years ago

#229 - fix

Pull Request - State: closed - Opened by JulSez about 4 years ago

#228 - feat(asymptotic): add asymptotic stationary experiment for rotting alg

Pull Request - State: closed - Opened by JulSez about 4 years ago

#227 - feat(delay): add utils for delay analysis

Pull Request - State: closed - Opened by JulSez about 4 years ago

#226 - feat(plot eff)

Pull Request - State: closed - Opened by JulSez about 4 years ago

#225 - Wrong implementation in Exp3++ algorithm

Issue - State: open - Opened by szrlee about 4 years ago - 4 comments

#224 - param(m) : change m param

Pull Request - State: closed - Opened by JulSez about 4 years ago

#223 - Experiment/eff

Pull Request - State: closed - Opened by JulSez over 4 years ago

#222 - fix(windows): exponential growth of the window

Pull Request - State: closed - Opened by JulSez over 4 years ago - 1 comment

#221 - set(new policy): m=1.5

Pull Request - State: closed - Opened by JulSez over 4 years ago - 1 comment

#220 - feat(new exp) : set new exp

Pull Request - State: closed - Opened by JulSez over 4 years ago

#219 - Link to todo file is broken

Issue - State: open - Opened by tomron over 4 years ago - 1 comment

#218 - feat(save): save time

Pull Request - State: closed - Opened by JulSez over 4 years ago

#217 - Seznec19b fig1

Pull Request - State: closed - Opened by JulSez over 4 years ago - 2 comments

#216 - Adds vendored Nix packaging.

Pull Request - State: closed - Opened by freuk over 4 years ago - 3 comments
Labels: enhancement, question

#215 - Seznec19a/fig3

Pull Request - State: closed - Opened by JulSez almost 5 years ago

#214 - fix(time) : time is init at zero

Pull Request - State: closed - Opened by JulSez almost 5 years ago - 1 comment

#213 - self.t = -1

Issue - State: closed - Opened by JulSez almost 5 years ago - 2 comments

#212 - Seznec19a/fig2

Pull Request - State: closed - Opened by JulSez almost 5 years ago - 1 comment
Labels: enhancement, single-player

#211 - Figure 1 : Seznec 19a

Pull Request - State: closed - Opened by JulSez almost 5 years ago

#210 - feat(rottingArm): add rested rotting arms

Pull Request - State: closed - Opened by JulSez almost 5 years ago
Labels: enhancement

#209 - Implement rotting arm

Issue - State: closed - Opened by JulSez almost 5 years ago - 4 comments
Labels: enhancement, single-player

#208 - Fixes for DUCB and kl-lcb computation

Pull Request - State: closed - Opened by JulSez almost 5 years ago - 3 comments

#207 - Implement TS-BU-Fin and TS-BU-Inf for piece-wise stationary bandits ?

Issue - State: open - Opened by Naereen almost 5 years ago - 2 comments
Labels: new algo, non-stationary

#206 - Rotting bandits policies and arms

Pull Request - State: closed - Opened by JulSez almost 5 years ago - 6 comments

#205 - Also submit the paper I wrote to JOSS?

Issue - State: open - Opened by Naereen almost 5 years ago
Labels: enhancement, documentation

#204 - Implement the improved GLR-klUCB independent of prior knowledge of Upsilon_T

Issue - State: closed - Opened by Naereen almost 5 years ago - 5 comments
Labels: enhancement, non-stationary

#203 - Try to use airspeed velocity to track benchmarks of this project?

Issue - State: closed - Opened by Naereen almost 5 years ago - 7 comments
Labels: enhancement, question, reports

#202 - Problem with scipy.optimize.root_scalar not found

Issue - State: closed - Opened by Naereen almost 5 years ago - 3 comments
Labels: bug

#201 - Implement the DPE (Decentralized Parsimonious Exploration) algorithm from arXiv:1909.13079

Issue - State: open - Opened by Naereen almost 5 years ago
Labels: enhancement, new algo, multi-player

#200 - Labels in boxplots are not displayed (again!)

Issue - State: closed - Opened by Naereen almost 5 years ago - 1 comment
Labels: bug

#198 - Publish SMPyBandits on Conda

Issue - State: open - Opened by Naereen about 5 years ago - 3 comments
Labels: enhancement, question

#197 - Implement basic simulation and algorithm for combinatorial bandit

Issue - State: open - Opened by Naereen about 5 years ago - 1 comment
Labels: enhancement, question, new algo, single-player

#196 - Implement more best arm identification algorithms

Issue - State: open - Opened by xuedong about 5 years ago - 3 comments
Labels: new algo, single-player

#195 - Update Thompson.py

Pull Request - State: closed - Opened by xuedong about 5 years ago - 1 comment
Labels: single-player

#194 - Try to use localization in change-point detectors

Issue - State: closed - Opened by Naereen over 5 years ago - 3 comments
Labels: enhancement, question, non-stationary

#193 - Update Gauss.py to correct #192

Pull Request - State: closed - Opened by Naereen over 5 years ago
Labels: bug

#192 - Gauss posterior

Issue - State: closed - Opened by janmotl over 5 years ago - 6 comments
Labels: bug, question

#191 - Use NEWMA algorithm as a change-point detector for non-stationary bandits from paper arXiv:1805.08061

Issue - State: open - Opened by Naereen over 5 years ago
Labels: enhancement, new algo, non-stationary

#190 - Try to not use any symbolic links (they don't work on Windows...)

Issue - State: open - Opened by Naereen over 5 years ago - 3 comments
Labels: bug

#189 - Make a short video on Windows showing installation, configuration and running experiments?

Issue - State: open - Opened by Naereen over 5 years ago - 1 comment
Labels: enhancement, question, documentation

#188 - Implement and add Exp3.S for switching bandits

Issue - State: closed - Opened by Naereen over 5 years ago - 3 comments
Labels: enhancement, new algo, non-stationary

#185 - Implement different arms means for different players for multi-player setting (arXiv:1902.01239)

Issue - State: open - Opened by Naereen over 5 years ago
Labels: enhancement, question, new algo, multi-player

#184 - Implement the C&P algorithm from arXiv:1902.08036

Issue - State: open - Opened by Naereen over 5 years ago
Labels: enhancement, new algo, multi-player, non-stationary

#183 - Add a clean support for piecewise stationary multi-player MAB problems

Issue - State: open - Opened by Naereen over 5 years ago - 3 comments
Labels: enhancement, multi-player, non-stationary

#182 - Implement the PHE algorithm from arXiv:1902.10089

Issue - State: closed - Opened by Naereen over 5 years ago - 3 comments
Labels: enhancement, new algo, single-player

#181 - Implement RCB algorithm from

Issue - State: closed - Opened by Naereen over 5 years ago - 5 comments
Labels: question, new algo, single-player

#180 - Store regret/time/memory/… (mean +- std) data for each experiments as LaTeX tables in small LaTeX files also

Issue - State: closed - Opened by Naereen over 5 years ago - 1 comment
Labels: enhancement

#179 - Better full reproductibility

Issue - State: closed - Opened by Naereen over 5 years ago - 1 comment
Labels: enhancement, multi-player, single-player

#178 - Plot nice histograms with seaborn.distplot instead of boxplot

Issue - State: closed - Opened by Naereen over 5 years ago - 2 comments
Labels: enhancement

#177 - Implement UCB-CDP and ImpCDP algorithms for non-stationary bandits

Issue - State: closed - Opened by Naereen almost 6 years ago - 1 comment
Labels: new algo, single-player, non-stationary

#176 - Implement the variant "with tracking" for GLR based algorithms

Issue - State: closed - Opened by Naereen almost 6 years ago - 1 comment
Labels: enhancement, non-stationary

#175 - Store, print and plot the mean/hist/boxplot of detection delay for CD based algorithms for non-stationary bandits

Issue - State: closed - Opened by Naereen almost 6 years ago - 4 comments
Labels: enhancement, question, non-stationary

#174 - Fix Monitored-UCB to reduce memory consumption

Issue - State: closed - Opened by Naereen almost 6 years ago
Labels: bug, non-stationary

#173 - Implement optimized variants of GLR tests

Issue - State: closed - Opened by Naereen almost 6 years ago - 4 comments
Labels: question, single-player, non-stationary

#172 - Implement a better oracle for non-stationary bandit

Issue - State: closed - Opened by Naereen almost 6 years ago
Labels: question, new algo, non-stationary

#171 - Use a smart default choice for alpha (proba_random_exploration) for CD algorithms

Issue - State: closed - Opened by Naereen almost 6 years ago - 1 comment
Labels: question, non-stationary

#170 - For CD_IndexPolicy algorithms, implement our variant of non-uniform forced exploration

Issue - State: closed - Opened by Naereen almost 6 years ago
Labels: new algo, non-stationary

#169 - Implement RR-SW-UCB# and SW-DLP from paper arXiv:1812.05165

Issue - State: open - Opened by Naereen almost 6 years ago
Labels: new algo, multi-player, non-stationary

#168 - Maybe there is an issue in how BaseWrapperPolicy restarts the underlying policy?

Issue - State: closed - Opened by Naereen almost 6 years ago - 1 comment
Labels: bug

#167 - Add a better legend for boxplot with lots of algorithms

Issue - State: closed - Opened by Naereen almost 6 years ago - 1 comment
Labels: bug

#166 - Wrong colors in histograms, don't know why

Issue - State: closed - Opened by Naereen almost 6 years ago - 2 comments
Labels: bug, enhancement, reports

#165 - Experiments in a notebook the statistical tests for non stationary bandit policies (CUSUM, PHT, GLR, Monitored etc)

Issue - State: closed - Opened by Naereen almost 6 years ago - 7 comments
Labels: enhancement, reports, non-stationary

#164 - Be sure that mean rewards are well scaled on the y axis for non stationary problems

Issue - State: closed - Opened by Naereen almost 6 years ago - 1 comment
Labels: bug, non-stationary

#163 - Be sure that last regrets and regrets can never be negative!

Issue - State: closed - Opened by Naereen almost 6 years ago - 2 comments
Labels: bug

#162 - Use Cython to speed up all the statistical tests for non stationary bandit policies (CUSUM, PHT, GLR, Monitored etc)

Issue - State: closed - Opened by Naereen almost 6 years ago - 3 comments
Labels: enhancement, question, non-stationary

#161 - Add a visualization: show last regrets as a boxplot (like time and memory) and not a collection of histograms

Issue - State: closed - Opened by Naereen almost 6 years ago - 5 comments
Labels: enhancement, single-player

#160 - Implement an oracle policy for the switching bandit problem, to have a strong baseline

Issue - State: closed - Opened by Naereen almost 6 years ago - 2 comments
Labels: new algo, non-stationary

#159 - Weird bug when using parallel computation: Tcl_AsyncDelete

Issue - State: closed - Opened by Naereen almost 6 years ago - 1 comment
Labels: bug, question

#158 - Increase figure size when saving and not displaying

Issue - State: closed - Opened by Naereen almost 6 years ago - 2 comments
Labels: enhancement

#157 - CD-MAB based on GLR tests for 1-dim exp. families (with kl)

Issue - State: closed - Opened by Naereen almost 6 years ago - 2 comments
Labels: new algo, single-player, non-stationary

#156 - Implement rotting bandit regret definition

Issue - State: closed - Opened by JulSez almost 6 years ago - 5 comments
Labels: enhancement, question, single-player

#153 - Clean up how I code non stationary bandits and especially the regret

Issue - State: closed - Opened by Naereen almost 6 years ago - 4 comments
Labels: bug, non-stationary

#151 - Implement LM-DSEE and SW-UCB# algorithms for non-stationary bandits from paper arXiv:1802.08380

Issue - State: closed - Opened by Naereen almost 6 years ago - 4 comments
Labels: new algo, single-player, non-stationary

#148 - Implement algorithm(s) from "Refining the Confidence Level for Optimistic Bandit Strategies" by T.Lattimore

Issue - State: open - Opened by Naereen almost 6 years ago
Labels: enhancement, new algo, single-player

#146 - Implement (environment and) algorithm(s) from "Adaptively Tracking the Best Arm with an Unknown Number of Distribution Changes"

Issue - State: closed - Opened by Naereen almost 6 years ago - 13 comments
Labels: enhancement, question, new algo, single-player, non-stationary

#140 - Implement algorithms from "An Optimal Algorithm for Stochastic and Adversarial Bandits", arXiv:1807.07623

Issue - State: closed - Opened by Naereen about 6 years ago - 10 comments
Labels: enhancement, question, new algo, single-player

#133 - Web based visualisation platform, locally

Issue - State: closed - Opened by Naereen over 6 years ago - 1 comment

#126 - Use https://github.com/Naereen/Kullback-Leibler-divergences-and-kl-UCB-indexes/ as git submodule

Issue - State: closed - Opened by Naereen over 6 years ago - 1 comment
Labels: enhancement

#117 - Add (basic) support for contextual bandits?

Issue - State: closed - Opened by Naereen over 6 years ago - 1 comment
Labels: enhancement, new algo, documentation

#100 - Implement EXP3.R to tackle switching bandits ?

Issue - State: closed - Opened by Naereen almost 7 years ago - 1 comment
Labels: new algo, single-player, non-stationary