Naren Sarayu Manoj

About me

I (Naren Sarayu Manoj) am a Quantitative Researcher at Citadel Fixed Income and Macro.

I received my PhD from the Toyota Technological Institute at Chicago, a philanthropically endowed CS research institution located on the University of Chicago campus with close ties to UChicago computer science. I was fortunate to be advised by Professors Avrim Blum and Yury Makarychev. From 2021-2024, I was generously supported by an NSF Graduate Research Fellowship. During summer 2023, I was a visiting student at EPFL hosted by Professor Michael Kapralov.

I graduated with my BS from the UT Austin computer science department in 2018.

Research interests

I study reliable machine learning—poisoning/backdoors, distributional robustness, and adversarial/semirandom structure. During my PhD, I approached these problems using tools from robust statistics, learning theory, and optimization. I’m also interested in resource-constrained settings like streaming and limited-feedback (e.g., dueling/bandit) optimization and what they can teach us about practical training and inference.

Publications

Most of my publications have authors ordered alphabetically, as is customary in theoretical computer science.

Distributionally Robust Linear Regression With Block Lewis Weights
Naren Sarayu Manoj, Kumar Kshitij Patel
ICLR 2026 (poster)
A preliminary version appeared at OPT 2024 workshop at NeurIPS 2024
The Change-of-Measure Method, Block Lewis Weights, and Approximating Matrix Block Norms
Naren Sarayu Manoj, Max Ovsiankin
SODA 2025 (arXiv) (poster) (slides)
On the Robustness of Spectral Algorithms for Semirandom Stochastic Block Models
Aditya Bhaskara, Agastya Vibhuti Jha, Michael Kapralov, Naren Sarayu Manoj, Davide Mazzali, Weronika Wrzos-Kaminska
NeurIPS 2024 (arXiv) (poster) (slides)
Near-Optimal Streaming Ellipsoidal Rounding for General Convex Polytopes
Yury Makarychev, Naren Sarayu Manoj, Max Ovsiankin
STOC 2024 (arXiv) (video) (slides)
Dueling Optimization with a Monotone Adversary
Avrim Blum, Meghal Gupta, Gene Li, Naren Sarayu Manoj, Aadirupa Saha, Yuanyuan Yang
ALT 2024 (Outstanding paper award) (arXiv) (poster) (slides)
A preliminary version appeared at OPT 2023 workshop at NeurIPS 2023 (Oral)
Interpolation Learning With Minimum Description Length
Naren Sarayu Manoj, Nathan Srebro
COLT 2023 (arXiv) (poster)
An Optimal Algorithm for Certifying Monotone Functions
Meghal Gupta, Naren Sarayu Manoj
SOSA 2023 (ECCC) (arXiv) (slides)
Streaming Algorithms for Ellipsoidal Approximation of Convex Polytopes
Yury Makarychev, Naren Sarayu Manoj, Max Ovsiankin
COLT 2022 (arXiv) (poster) (slides)
Excess Capacity and Backdoor Poisoning
Avrim Blum, Naren Sarayu Manoj
NeurIPS 2021 (Spotlight) (arXiv) (slides) (code)
Development and Validation of a Deep Learning Algorithm for Gleason Grading of Prostate Cancer from Biopsy Specimens
Kunal Nagpal, Davis Foote, ..., Naren Sarayu Manoj, ...,
JAMA Oncology (journal)
Random Smoothing Might be Unable to Certify $\ell_\infty$ Robustness for High-Dimensional Images
Avrim Blum, Travis Dick, Naren Sarayu Manoj, Hongyang Zhang
JMLR 2020 (arXiv)

Preprints

Quantifying Perceptual Distortion of Adversarial Examples
Matt Jordan, Naren Sarayu Manoj, Surbhi Goel, Alexandros G. Dimakis
arXiv 2019 (arXiv)

Reviewing

AISTATS, Annals of Statistics, FOCS, ICALP, NeurIPS, STOC

For fun

I play exceedingly amateurish chess.