Ahmed Hefny Photo

Ahmed Hefny

Machine Learning Department
Carnegie Mellon University

Office: GHC 8223
E-mail: ahefny AT CSdotCMU DOT edu
Home | Publications | Resume | Misc

Welcome to my web-page. I am a PhD candidate in the Machine Learning Department. I am currently working with Prof. Geoffrey Gordon. My current research focuses on building better models for learning and control of dynamical systems. My goal is to provide practical methods with provable guarantees for learning to act and predict in partially observable environments. My research towards this goal draws elements and ideas from predictive state representations, kernel methods, recurrent neural networks and reinforcement learning. My other interests include optimization, building probabilistic models for textual and social data, and software engineering for machine learning.

Prior to joining CMU, I got my BSc and MSc in computer engineering from Cairo University, Egypt. My MSc thesis was supervised by Prof. Amir Atiya. While working as a teaching assistant at Cairo University, I was also an R&D engineer at IBM Cairo Technology Development Center and then a research assistant at Cairo Microsoft Innovation Center (currently known as Advanced Technology Labs Cairo) with Dr Kareem Darwish.

News

  • I gave a talk at the UC Berkley Center for Human-Compatible AI.
  • Our team won the first place in the first neurohackathon hosted by CMU BrainHub.
  • Our paper on predictive belief methods for learning dynamical systems has been accepted in NIPS 2015. (preprint)
  • Our paper on asynchronous variants of variance-reduced SGD has been accepted in NIPS 2015.
  • I spent summer 2015 in Google[x], working on behavior prediction for the self-driving car .
  • Our paper on large scale coordinate-descent with linear coupling constrains has been accepted in UAI 2015.
  • I am TAing spring 2015 offering of 10601 (Machine Learning) with Tom Mitchell and Nina Balcan.
  • I spent summer 2014 in Google Mountain View as a software engineering intern, building models for user prediction.
  • I am TAing fall 2013 offering of 10701 (Machine Learning) with Geoff Gordon and Alex Smola.
  • I spent summer 2013 in a Bing/MSR joint internship in Redmond.
  • My paper with Avinava Dubey, Sinead Williamson and Eric Xing on non-parametric topic modeling over time has been accepted in SIAM Datamining Conference 2013.
  • I spent summer 2012 in Bellevue as a Microsoft Research intern, working with Bing Document Understanding team.

Coursework

  • Machine Learning (Eric Xing)
  • Intermediate Statistics (Larry Wasserman)
  • Statistical Machine Learning (Larry Wasserman)
  • Graduate Algorithms (Manuel Blum)
  • Optimization (Geoffrey Gordon and Ryan Tibshirani)
  • Multimedia Databases and Data Mining (Christos Faloutsos)
  • Advanced Probability Overview (Jing Lei)
  • Probabilistic Graphical Models (Eric Xing)
  • Spectral Graph Theory (Gary Miller)
  • Deep Learning (Bhishka Raj)
  • Randomized Algorithms and Advanced Optimization (Alexander Smola and Suvrit Sra)
  • Topics in Deep Learning (Ruslan Salakhutdinov)

Software




Last update: Aug 14, 2017