Fadi Atieh

I am a PhD student in Machine Learning at Carnegie Mellon University, advised by Max Simchowitz and Nicholas Boffi. I hold a B.S. in Mathematics and Computer Science (Courses 18 & 6-3) and an M.Eng. in EECS from MIT. My master’s thesis, supervised by Yury Polyanskiy, developed a novel statistical hypothesis testing procedure for the discovery of the Higgs boson using LHC data.

Prior to my PhD, I spent approximately three years as a Quantitative Research Engineer in Global Quantitative Strategies (GQS) at Citadel. During my non-compete period, I collaborated remotely with the group of Professor Aditi Krishnapriyan at UC Berkeley, leading to co-authoring the paper Transformers Discover Molecular Structure Without Graph Priors.

My research interests lie at the intersection of generative modeling, reinforcement learning, robotics, and applied mathematics. I am supported by the DOE Computational Science Graduate Fellowship (CSGF).

I was born and raised in Syria and moved to the United States for college.

Email / CV / Scholar / Github

news

Aug 24, 2026	Excited to share that I joined Amazon Frontier AI & Robotics (FAR) as an Applied Scientist Intern, starting August 24, 2026!
Apr 23, 2026	Honored to receive the DOE Computational Science Graduate Fellowship (CSGF)!
Apr 14, 2026	Excited to announce that I will be joining the Machine Learning Department at Carnegie Mellon University as a PhD student starting Fall 2026!

publications

2025

Transformers Discover Molecular Structure Without Graph Priors

T. Kreiman, Y. Bai, F. Atieh, and 3 more authors

2025

arXiv Bib

@article{kreiman2025transformers,
  title = {Transformers Discover Molecular Structure Without Graph Priors},
  author = {Kreiman, T. and Bai, Y. and Atieh, F. and Weaver, E. and Qu, E. and Krishnapriyan, A. S.},
  year = {2025},
}

2020

Heuristic Recurrent Algorithms for Photonic Ising Machines

C. Roques-Carmes, Y. Shen, C. Zanoci, and 10 more authors

Nature Communications, 2020

arXiv Bib HTML

@article{roquescarmes2020heuristic,
  title = {Heuristic Recurrent Algorithms for Photonic Ising Machines},
  author = {Roques-Carmes, C. and Shen, Y. and Zanoci, C. and Prabhu, M. and Atieh, F. and Jing, L. and Dub\v{c}ek, T. and Mao, C. and Johnson, M. R. and \v{C}epri\'{c}, V. and Joannopoulos, J. D. and Englund, D. and Solja\v{c}i\'{c}, M.},
  journal = {Nature Communications},
  year = {2020},
}

notes

Working notes where I elaborate on concepts I'm currently studying. When a proof or technique seems to come out of the blue, I try to reverse-engineer how one might have arrived at it, reconstructing the chain of reasoning from scratch.

On the Contractivity of Stochastic Interpolation Flow — August 15, 2026
How strong log-concavity of the target controls the Lipschitz constant of the interpolant ODE flow map.

From Behavior Cloning to RL — August 10, 2026
Tracing the arc from vanilla BC through DAgger and AggreVaTe to approximate policy iteration.

How much of an approximation is PPO? — July 25, 2026
Relating the discounted return to the unclipped PPO surrogate via state occupancy measures and the performance difference lemma, and clarifying which additional approximations hold in practice.

Vector-Time Flows — July 07, 2026
Generalizing stochastic interpolants to per-coordinate time schedules, and recovering in-painting as a special case.

Optimality Conditions in Finite and Infinite Dimensions — May 16, 2026
Deriving necessary and sufficient conditions for local optima, from unconstrained and constrained finite-dimensional problems to infinite-dimensional optimization.

View all notes →