Piyushi Manupriya
About
I am an ML postdoc, immensely fortunate to be working with Prof. Anant Raj on optimization & ML theory.
Prior to this, I finished my M.Tech-PhD dual degree (Aug '18-Mar '25) from CSE IIT Hyderabad, during which I worked on research problems in Optimal Transport,
Kernel Methods, Bandits and on Deep Learning Interpretability.
ArXived
Publications
Optimization & Online Learning
Multi-agent Multi-armed Bandits with Minimum Reward Guarantee Fairness
P Manupriya, Himanshu, JS Nath, G Ghalme
AAMAS 2025 (CORE A*) as an Oral
Optimal Transport & Kernels
MMD-regularized unbalanced optimal transport
P Manupriya, JS Nath, P Jawanpuria
Transactions on Machine Learning Research, 2024
Among the 100 papers invited at ICLR 2025 (CORE A*)
Submodular framework for structured-sparse optimal transport
P Manupriya, P Jawanpuria, KS Gurumoorthy, JS Nath, B Mishra
ICML 2024 (CORE A*)
Consistent optimal transport with empirical conditional measures
P Manupriya, RK Das, S Biswas, JS Nath
AISTATS 2024 (CORE A)
Deep Learning Interpretability
Neural network attributions: a causal perspective
A Chattopadhyay, P Manupriya, A Sarkar, VN Balasubramanian
ICML 2019 (CORE A*) as an Oral
Improving attribution methods by learning submodular functions
P Manupriya, TR Menta, JS Nath, VN Balasubramanian
AISTATS 2022 (CORE A)
Research Grants Availed
- ANRF-National Post Doctoral Fellowship (NPDF) grant
- 50K USD Early-Stage Google PhD Fellowship in Machine Learning
- Ministry of Education (MoE) PhD Fellowship
Teaching Assistant Roles at IIT Hyderabad
AI5030 Probability and Stochastic Processes (Fall '24)
Role: Tutorials (theory), grading quizzes, assignments & exams
Course Instructor: Prof. P.N. Karthik
CS5590 Foundations of ML (Fall '22)
Role: Tutorials (theory & coding), grading quizzes & assignments
Course Instructor: Prof. J. Saketha Nath
CS5580 Convex Optimization Theory (Fall '21 & '23)
Role: Grading quizzes
Course Instructor: Prof. J. Saketha Nath
CS5560 Probabilistic Models for ML (Spring '21 & '22)
Role: Grading quizzes
Course Instructor: Prof. J. Saketha Nath
CS5350 Bayesian Data Analysis (Spring '24)
Role: Grading exams
Course Instructor: Prof. Srijith P.K.
CS6510 Applied ML (Fall '19 & '20)
Role: Grading quizzes & exams
Course Instructor: Prof. Vineeth N Balasubramanian
Other
Maths tutor (Dec '21 - Mar '22) for a batch of 3 Class-X students at the Aksharamaala Program.
Mentoring
Gautham Bellamkonda (IIT-H BTech CSE '24 batch)
Vanshika Mittal (IISc Predoc in Prof Anant's lab)
G. Y. Srikrishna (IISc PhD student in Prof Anant's lab)
Services
Reviewer at ICML '26, ISIT '25, AISTATS '24
Sub-reviewer at NeurIPS '23, NeurIPS '20, ICML '20, ECCV '20, WACV '20, ICCV '19, ECML-PKDD '19
Volunteer: Women in ML (WiML) symposium at ICML '24; for NeurIPS ('20 & '21) and ICML '20 virtual conferences
News
- Jan '26 Presented our AAMAS '25 paper as an invited (Best Paper Finalist) talk at exSPLOre '26
- Sep '25 Our proposal on optimization received the ANRF-NPDF grant
- Aug '25 Presented poster for our AAMAS '25 paper at the ICTS event
- Jul '25 Attended the BIRS-CMI Game-Theoretic Statistical Inference workshop. Grateful to Prof. Shubhada Agrawal & Prof. Aaditya Ramdas for the opportunity.
- Jun '25 Presented my PhD work on Optimal Transport at IIT Madras. Thanks to Prof. Harish Guruprasad for the kind invitation.
- Apr '25 Presented our TMLR '24 work as an invited poster at ICLR '25
- Mar '25 PhD Thesis, Improving Optimal Transport using Kernel Methods, released on HAL. BibTeX.
- Mar '25 Defended my thesis (slides)
- Mar '25 Started postdoc work in Prof. Anant's lab
- Feb '25 Presented our ICML '24 work as an invited poster at ACM ARCS '25
- Dec '24 Multi-agent Multi-armed Bandits with Minimum Reward Guarantee Fairness (link) got accepted at AAMAS '25 (CORE A*) as an Oral
- Dec '24 Submitted my PhD thesis and gave the open colloquium at IIT Hyderabad
- Nov '24 My exploratory observations on fairness in matchings will be presented as an extended abstract at AFME, NeurIPS '24 workshop.
- Jul '24 Presented our ICML '24 work at the conference in Vienna, Austria
- May '24 Submodular Framework for Structured-Sparse OT (link) got accepted at ICML '24 (CORE A*)
- Jan '24 Consistent OT with Empirical Conditional Measures (link) got accepted at AISTATS '24 (CORE A)
- Dec '23 MMD-Regularized OT (link) accepted at TMLR Journal
- Sep '23 Attended the global Google PhD Fellowship summit (online)
- Jan '23 Attended the Research Week with Google
- Jul '22 Attended the global Google PhD Fellowship summit (online)
- Mar '22 Presented our AISTATS '22 work at the virtual conference
- Jan '22 My M.Tech project on ensembling saliency maps using submodular functions (link) got accepted at AISTATS '22 (CORE A).
- Oct '21 Got selected for the PMRF Fellowship (courteously declined for Google's)
- Sep '21 Received the early-stage Google PhD Fellowship in Machine Learning.
- Aug '21 Cleared the Comprehensive exam. Chosen seminar paper: Scaling Algorithms for Unbalanced Optimal Transport Problems (Chizat et al., 2017).
- Aug '20 Converted from MTech-Research to PhD at IIT Hyderabad
- May '20 Submodular Ensembled Attribution got accepted at ICML Workshops (XXAI - Poster; WHI - Spotlight).
- May '19 Neural Network Attributions: a Causal Perspective (link) got accepted at ICML '19 (CORE A*) as an Oral