Email: the first letter of my first name followed by my last name at

Curriculum Vitae 

Research Interests

My dear friend Robert West once called me a "true computer scientist." I was honestly a bit puzzled at the time since my research heavily features statistics, optimization, and computational mathematics in addition to algorithms; my guilty pleasure is finding reasons to learn and incorporate all sorts of math into my work. However, at the core of it, computer science is also a combination of disciplines with a strong lean towards elucidating mathematical structure. Oftentimes such structure is the centerpiece of my work - I won a best thesis award for my PhD's exploration of the combinatorial structure underlying text learning problems. My research interests are constantly evolving as I learn about new areas, and I am currently focused on

  1. connecting the deep and statistical learning worlds by designing learning paradigms that leverage the strengths of both,

  2. developing novel learning methodologies for science, particularly ones that can effectively incorporate or be analyzed by scientific theories,

  3. novel optimization techniques based on convex algebraic geometry.

While seemingly disparate, these interests cater to my insatiable desire to push the boundaries of what is computationally feasible and to explore problems that are not clearly solvable.

Upcoming Publications

Publications by Topic

Learning via Compression

Large-scale Learning​ Algorithms




  • Learning with N-Grams: from Massive Scales to Compressed Representations. Hristo Paskov. Stanford Computer Science PhD Thesis, 2016. Awarded Arthur Samuel Award for Best Dissertation in Computer Science at Stanford.

  • A Regularization Framework for Active Learning from Imbalanced Data. Hristo Paskov. MIT Master's Thesis, 2010.