KL-UCB #59
Closed
KL-UCB #59
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This comment was marked as resolved.
This comment was marked as resolved.
shashist
reviewed
Aug 29, 2023
It does, as a child class of |
This comment was marked as resolved.
This comment was marked as resolved.
shashist
approved these changes
Sep 1, 2023
There was a problem hiding this comment.
Code part is ok. @Arkadiy-Vladimirov please sync this PR with recent changes and fix CI
Add BayesUCB to DDPG and accelerated the speed of learning See merge request ai-lab-pmo/mltools/recsys/RePlay!94
[CI] Running CI in CI_DEFAULT_BRANCH and publish dev packages See merge request ai-lab-pmo/mltools/recsys/RePlay!103
Fix package registry path See merge request ai-lab-pmo/mltools/recsys/RePlay!104
Move some functionality to experimental part See merge request ai-lab-pmo/mltools/recsys/RePlay!93
Add dataset functionality See merge request ai-lab-pmo/mltools/recsys/RePlay!102
[Feature] Building experimental package See merge request ai-lab-pmo/mltools/recsys/RePlay!108
Add new splitters See merge request ai-lab-pmo/mltools/recsys/RePlay!105
Feature/add new preprocessing See merge request ai-lab-pmo/mltools/recsys/RePlay!111
Update metrics interfaces See merge request ai-lab-pmo/mltools/recsys/RePlay!112
Fix metrics See merge request ai-lab-pmo/mltools/recsys/RePlay!114
Rename experiments to examples and add testing for notebooks (disabled to save resources) See merge request ai-lab-pmo/mltools/recsys/RePlay!122
Update CI 0.0.7 See merge request ai-lab-pmo/mltools/recsys/RePlay!123
merge v0.13.0 commit to main See merge request ai-lab-pmo/mltools/recsys/RePlay!126
Little fixes in toml See merge request ai-lab-pmo/mltools/recsys/RePlay!125
|
Please sync with main and I will rerun CI |
levensons
force-pushed
the
feature/kl_ucb
branch
from
dae0dfc to
62cb686
Compare
bysheva
reviewed
Nov 23, 2023
bysheva
requested changes
Nov 23, 2023
|
Tests should be fixed. Currently they will result in |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The PR proposes
KL-UCBrecommender based on the article by Aurélien Garivier and Olivier Cappé. As a recommender modelKL-UCBoperates just as the regularUCBwith the only difference in upper bound computation. This allowed to implementKL-UCBas a subclass ofUCBwith basically the only_calc_item_popularitymethod overridden - one that calculates the bound.__init__method is slightly changed only to provide more adequate default parameterexploration_coef = 0as it was claimed to be the most suitable in practice by the authors of the article. You may read more information onKL-UCBwhen recompile the documentation or in the class docstring directly. Evaluation is done in02_model_comparison.ipynbnotebook. You may take a look at metrics there or in the picture below.