A
Andrej Risteski
Researcher at Carnegie Mellon University
Publications - 76
Citations - 1579
Andrej Risteski is an academic researcher from Carnegie Mellon University. The author has contributed to research in topics: Computer science & Word (computer architecture). The author has an hindex of 18, co-authored 64 publications receiving 1208 citations. Previous affiliations of Andrej Risteski include Massachusetts Institute of Technology & Princeton University.
Papers
More filters
Journal ArticleDOI
A Latent Variable Model Approach to PMI-based Word Embeddings
TL;DR: A new generative model is proposed, a dynamic version of the log-linear topic model of Mnih and Hinton (2007) to use the prior to compute closed form expressions for word statistics, and it is shown that latent word vectors are fairly uniformly dispersed in space.
Journal ArticleDOI
Linear Algebraic Structure of Word Senses, with Applications to Polysemy
TL;DR: This article showed that multiple word senses reside in linear super-supervised super-embeddings for polysemous word embeddings and showed that the representations of multiple senses of a word are similar.
Proceedings Article
Do GANs learn the distribution? Some Theory and Empirics
TL;DR: In this paper, the authors proposed a novel test for estimating support size using the birthday paradox of discrete probability, and theoretically studied encoder-decoder GANs architectures (e.g., BiGAN/ALI), which were proposed to learn more meaningful features via GAN and consequently to also solve the mode-collapse issue.
Posted Content
The Risks of Invariant Risk Minimization
TL;DR: In this setting, the first analysis of classification under the IRM objective is presented, and it is found that IRM and its alternatives fundamentally do not improve over standard Empirical Risk Minimization.
Posted Content
Random Walks on Context Spaces: Towards an Explanation of the Mysteries of Semantic Word Embeddings.
TL;DR: A rigorous mathematical analysis is performed using the model priors to arrive at a simple closed form expression that approximately relates co-occurrence statistics and word embeddings, and leads to good solutions to analogy tasks.