Soheil Sayah Varg
B.Sc. Computer Engineering, Sharif University of Technology
Tehran, Iran
I work on inference and approximation under uncertainty: when the thing you care about cannot be computed exactly — a posterior, a value function, an expectation over an intractable distribution — you compute an approximation instead, and the question that interests me is what licenses trusting it.
01Research
Exact controlled generation from tilted autoregressive targets
Led the research and algorithm design for sampling from reward-tilted autoregressive language models using a particle-Gibbs ladder over a batched proposal chain. Worked on correctness and target-invariance arguments, implemented the method, and ran its experiments. First-author manuscript in preparation.
Model-based reinforcement learning in confounded POMDPs
Led a three-person project implementing and evaluating the method of Hong, Qi, and Xu (ICML 2024). Built the experimental pipeline, implemented bridge-function identification using negative-control proxies, validated it against an exact oracle, and designed experiments to examine its finite-sample behaviour. Took responsibility for writing the final results. Public code and results.
Bandit literature review
Studying batched bandits, privacy-constrained linear bandits, and misspecified kernelized bandit optimization.
02Selected projects
model-based-rl-confounded-pomdps
End-to-end empirical implementation of model-based RL for confounded POMDPs, with exact-oracle checks and a documented pessimism failure mode.
war-of-attrition-rl
Bayesian type filtering and potential-based reward shaping for a multi-stage war of attrition with incomplete information.
modern-information-retrieval
Dense + sparse retrieval fused by RRF, with cross-encoder reranking.
convex-optimization-algorithms
Chambolle-Pock and ADMM written from their update rules, checked against a solver rather than delegated to one.
market-predictability-study
A backtest of mine returned several hundred percent. This is the study of whether any of the reasons I gave for it hold up. They do not.
segmentation-and-deep-rl
Attention U-Net variants for segmentation, and Soft Actor-Critic built from scratch in PyTorch.
neogit
A version control system written from scratch in C — staging, commits, branches, checkout, revert, tags, and a pre-commit hook framework. No libraries.
imdb-postgresql-optimization
~190M IMDb rows into PostgreSQL via a SKIP LOCKED queue, then eight queries indexed with real before/after measurements.
03Teaching
Teaching Assistant, Sharif University of Technology
Multi-Agent Reinforcement Learning (MARL, ×1); Machine Learning (×2); Deep Learning (×1); High-Dimensional Statistics (×1); Stochastic Processes (×2); Probability & Statistics (×3); Artificial Intelligence (×1); Data Science (×1); Signals & Systems (×1, Head TA); Linear Algebra (×3); Data Structures & Algorithms (×1); Logic Design (×1); Fundamentals of Programming (×1). Have begun preparing theoretical questions for the upcoming term.
Co-founder, Resonance
An online Physics Olympiad school, started as a small study group, for students who have no selective high school near them. I teach the mechanics, electromagnetics and laboratory tracks and write problem sets for the national selection rounds.
04Education and honors
B.Sc. Computer Engineering, Sharif University of Technology
GPA 19.45 / 20.0, ranked 15th of 185 — the departmental average for this cohort is 17.45. Stochastic Processes and Reinforcement Learning taken as graduate courses.
Gold Medal, Iranian National Physics Olympiad — 2023
7th nationwide among more than 10,000 participants, and 2nd nationwide in the experimental examination. Admitted to Sharif through the olympiad medalist track.
Juror, 13th Persian Young Naturalists’ Tournament — 2025
An IYPT-format research tournament under the World Federation of Physics Competitions. Judged team defenses of open-ended physics research problems.