Tomek Korbak's repositories
pretraining-with-human-feedback
Code accompanying the paper Pretraining Language Models with Human Preferences
active-inference
A toy model of Friston's active inference in Tensorflow
treehopper
A Tree-LSTM-based dependency tree sentiment labeler
measuring-non-trivial-compositionality
Code accompanying the paper Measuring non-trivial compositionality in emergent communication
quora-question-pairs
This repo contains models by the team of @360er0 and @tomekkorbak as they participated in Quora question pairs Kaggle contest.
lewis-signaling-games-in-python
Code accompanying the blog post "Introduction to Lewis signaling games with Python"
nkjp-tagger
A tensorflow window-based POS tagger for National Corpus of Polish
cifar10-experiments
I explore CIFAR-10 dataset, estabilish a kNN benchmark, extract features using a pre-trained ConvNet and then fit and evaluate an SVM classifier
spinningup
An educational resource to help anyone learn deep reinforcement learning.
compositional-communication-via-template-transfer
Code acompanying the paper Developmentally motivated emergence of compositional communication via template transfer
awesome-polish-nlp
Resources for doing NLP in Polish
awesome-RLHF
A curated list of reinforcement learning with human feedback resources (continually updated)
computational-genomics-experiments
A couple of assignments I did for a course in computational (molecular) biology @ University of Warsaw
entropy-of-programming-languages
Estimating information entropy rates across programming languagues
parallel-b-suitors
A parallel C++ implementation of b-suitors algorithm for solving the b-matching problem
prodigy-recipes
✨ Recipes for the Prodigy annotation tool
Sussex_PhDThesis
Template for a PhD thesis at the University of Sussex