Khashayar Gatmiry
Scholar

Khashayar Gatmiry

Google Scholar ID: w8XocYQAAAAJ
Graduate student, Massachusetts Institute of Technology
Machine LearningSamplingOptimization
Citations & Impact
All-time
Citations
320
 
H-index
11
 
i10-index
11
 
Publications
20
 
Co-authors
19
list available
Resume (English only)
Academic Achievements
  • Published multiple papers covering topics such as transformers, implicit bias, algorithmic generalization and stability, diffusion models, MCMC sampling, online learning/optimization, with some appearing in top conferences like ICML 2024 and NeuRIPS 2023.
Research Experience
  • In Fall 2023, a research intern at Google, NYC, in the BigML group, supervised by Sashank Reddi and Sobhan Miryouseffi, exploring methods to improve the training time of the Bert model; previously a research intern at Microsoft in the Foundation of Machine Learning group, working on the capabilities of small language models for building an embedding system for short stories, advised by Ronen Eldan, Adil Salim, and Yi Zhang.
Education
  • Massachusetts Institute of Technology (MIT), Electrical Engineering and Computer Science (EECS), advised by Stefanie Jegelka and Jonathan Kelner.
Background
  • A final-year graduate student at MIT, EECS, and CSAIL. Research interests include deep learning, sampling, and optimization.