Artyom Gadetsky

I am 2nd year PhD student at MLBio lab, EPFL in Lausanne, supervised by Maria Brbic. Currently, I am working on unsupervised transfer and representation learning. Previously, I was a researcher at CS HSE and BayesGroup supervised by Dmitry Vetrov, working on stochastic optimization w.r.t. discrete distributions over structured objects such as graphs, permutations, etc. Generally speaking, I am interested in different applications of probabilistic modelling to solve general machine learning problems as well as real-life problems.

Email  /  CV  /  Google Scholar  /  Github  /  Twitter  /  LinkedIn

An approach to perform fully unsupervised transfer on a downstream dataset given representation spaces of foundation models. Although being fully unsupervised, our approach outperforms CLIP zero-shot transfer and sometimes matches optimal supervised performance!

We develop the approach to infer fine-grained labels given coarsely labeled dataset.

Simple model-agnostic framework for inferring human labeling of a given dataset without any external supervision.

Tractable discrete distributions and stochastic optimization w.r.t. discrete distributions over structured objects (i.e. graphs, permutations, top-k).

Variance reduction techniques for stochastic optimization w.r.t. a distribution over permutations.

The generative model for definitions of polysemous words given vector representation of the word and example of its use.