Welcome to Rishabh Iyer's webpage
Assistant Professor, University of Texas Dallas
I am currently an Assistant Professor at the University of Texas, Dallas, where I lead the CARAML Lab. I'm also a Visiting Assistant Professor at the Indian Institute of Technology, Bombay. Before this, I was a Senior Research Scientist at Microsoft from 2016 - 2019. Below are some of the research topics my group is currently actively working on:
Combinatorial Loss Functions: A new family of Combinatorially inspired loss functions for representation learning and self-supervised learning problems: SCORE (ICML 2024), SMILE (ECCV 2024)
Compute-Efficient Learning via Data Subset Selection and Coresets: We developed approaches which have obtained 5x - 10x speedups/energy efficiency with small data subsets with negligible loss in accuracy (generalization performance) applied to varied applications like supervised, semi-supervised, domain adaptation, hyper-parameter tuning, NAS, Federated Learning and Large Language Model Finetuning. We have applied these methods on a number of domains like image classification, NLP, speech recognition - GLISTER (AAAI 2021), GRADMATCH (ICML 2021), RETRIEVE (NeurIPS 2021), SELCON (ICML 2021), ORIENT (NeurIPS 2022), AUTOMATA (NeurIPS 2022), SubSelNet (NeurIPS 2023), PGM (EMNLP 2022), INGENIOUS (EMNLP 2023), GCFL (WACV 2024)
Active Learning: Learning with fewer labels, reducing labeling costs by 2x - 5x (see this blog page on some of our work in active learning), particularly in realistic scenarios like rare classes/slices, imbalance, OOD instances, redundancy, etc. We have applied this on various problems including image classification, object detection, segmentation, text classification and NLP, and 3D Object Detection - FASS (ICML 2015), Learning with Less Data (WACV 2019), SIMILAR (NeurIPS 2021), TALISMAN (ECCV 2022), PRISM (AAAI 2022), CLINICAL (MICCAI 2022), DIAGNOSE (MICCAI 2022), DITTO (ACL 2023), CLARIFIER (WACV 2024), STONE (NeurIPS 2024).
Robust Deep Learning in the presence of outliers, label noise, out of distribution data, and so on - WRSSL (ICDM 2022), NestedMAML (AAAI 2022)
Data Programming and Weak Supervision, Learning with Rules, Labeling Functions - SPEAR (ACL 2021), WISDOM (ACL 2022), RuleAugmentedCP (ACL 2021)
Other Areas: Continuous Learning - GCR (CVPR 2022), Semi-Supervised Meta Learning - PLATINUM (ICML 2022)
Discrete Optimization (specifically submodular optimization), Combinatorial (Submodular) Information Measures - See My PhD Thesis, The SMI Journal Paper
Our research is currently supported by grants from NSF, Adobe Data Science Award, a Google Gift, Amazon Research Award, and the UT Dallas Seed grant. Thank you! Our research is motivated by real-world problems in machine learning, computer vision, text, and NLP! For more on my research, please see my research page, my publications, or my lab webpage.
I completed my Ph.D. in 2015 from the University of Washington, Seattle where I worked with Jeff Bilmes. I am excited about making machines assist humans in processing massive amounts of data, particularly in understanding videos and images. I am interested in building intelligent systems which organize, analyze and summarize massive amounts of data, and also automatically learn from this.
I received the best paper awards at Neural Information Processing Systems (NeurIPS/NIPS) in 2013, the International Conference of Machine Learning (ICML) in 2013, and an Honorable Mention at CODS-COMAD in 2021. I also received several research awards including an NSF Medium Grant, an Adobe Data Science Research Award, Microsoft Research Ph.D. Fellowship, Facebook Ph.D. Fellowship, and the Yang Award for Outstanding Graduate Student from the University of Washington.
For more information, please see my Google Scholar Profile, LinkedIn Profile, DBLP, or my GitHub page. I also maintain a YouTube channel where I add videos of my lectures and research talks.
Twitter: @rishiyer
Awards and Recognition
Amazon Research Award on Fairness for AI, July 2022 (Main PI: Preethi Jyothi)
Adobe Data Science Research Award, February 2022
NSF Medium Grant, September 2021.
Honorable Mention for our paper at CODS-COMAD 2021
Outstanding Reviewer Award for NeurIPS 2020 and 2021!
Finalist in the LDV Computer Vision Conference, New York in 2017
Yang Outstanding Graduate Student Award, University of Washington, Seattle
Microsoft Research Fellowship Award, 2014
Facebook Fellowship Award. 2014 (Declined in favor of Microsoft)
Best Paper Award at the International Conference of Machine Learning, 2013
Best Paper Award at the Neural Information Processing Systems Conference, 2013
Work Experience and Education
Spring 2020 to Present, Assistant Professor at the CS Department, UT Dallas
August 2020 to Present, Visiting Assistant Professor at CSE Department, IIT Bombay
March 2016 - December 2019, Senior Research Scientist, Microsoft
March 2015 - March 2016, Post-Doctoral Researcher, University of Washington
September 2011 - March 2015, M.S and Ph.D., University of Washington, Seattle
August 2011 - May 2011, B.Tech, IIT-Bombay
Teaching
University of Texas at Dallas
Introduction to Machine Learning (Undergraduate Course): Spring 2022, Spring 2024
Machine Learning: Fall 2020, Fall 2021, Fall 2022, Spring 2023, Fall 2023 (Course Website, Youtube Playlist of Video Lectures)
Advanced Optimization in Machine Learning (Course Website, Youtube Playlist of Video Lectures): Spring 2021
Optimization in Machine Learning (Course Website): Spring 2020
University of Washington
Spring 2014: Teaching Assistant for Submodular Functions, Optimization, and Applications to Machine Learning
Fall 2011: Introduction to Electrical Engineering
Indian Institute of Technology Bombay
Summer 2015: Instructor at the Mini-course on Submodular Optimization at the Non-Convex Optimization in Machine Learning, IIT Bombay (Slides)
Spring 2010: Teaching Assistant for Introduction to Probability and Statistics
Tutorials/Workshops at Conferences
AAAI 2022 (tutorial website)
SubSetML-2021 at ICML 2021 (workshop website)
IJCAI-PRICAI 2020 (tutorial website)
ECAI 2020 (tutorial website)
WACV 2019 (tutorial website)
Recent news
Serving as an Area Chair for ICLR 2025 and AAAI 2025.
STONE (An Active Learning Framework for 3D Object Detection) is accepted at NeurIPS 2024!
SMILE (Submodular Mutual Information Based Loss Functions for Few Shot Object Detection) is accepted at ECCV 2024!
SCORE (Submodular Combinatorial Representation Learning) is accepted at ICML 2024! We introduce a new family of loss functions for representations learning motivated by submodular functions!
Served as a Panelist for NSF Medium Proposals, April 2024.
Serving as an Area Chair for ICLR 2024 and AAAI 2024!
CLARIFIER and GCFL accepted at WACV 2024!
SubSelNet accepted at NeurIPS 2023!
INGENIOUS (compute-efficient LLM pre-training) accepted at Findings of EMNLP 2023!
DITTO (targeted subset selection for ASR Accent adaptation) is accepted at ACL 2023!
PRESTO on Mixed Discrete Continuous Optimization for Mixture Modeling is accepted at ICML 2023!
Joined the Editorial Board of Transactions of Pattern Analysis and Machine Intelligence (TPAMI) as an Action Editor!
Invited talk at IndoML 2022 (video here) on Subset Selection for Human in the Loop ML.
Two papers: ORIENT on data subset selection for distribution shift, and AUTOMATA on gradient-based subset selection for compute-efficient hyper-parameter tuning are accepted at NeurIPS 2022!
Our work on Robust Semi-supervised Learning was accepted at ICDM 2022! Congrats Krishnateja and Xujiang!
Excited to Recieve an Amazon Research Award in Alexa Fairness for AI track for our proposal on Fair Speech Recognition using Targeted Subset Selection and Active Semi-supervised Learning (joint with Preethi Jyothi and Ganesh Ramakrishnan from IIT Bombay).
Our work TALISMAN, which is on Targeted Active Learning for Object Detection in Autonomous Driving with Rare Scenarios (e.g, Mining instances of "Motorcycles at night" or "Pedestrians on a Highway in Foggy Scenarios") accepted at ECCV 2022!
Our work on Semi-supervised Meta-Learning (PLATINUM) was accepted to ICML 2022!
Excited to receive the Adobe Data Science Research Award for February 2022! Thanks, Adobe for the support!
Our work on gradient coresets for continuous learning was accepted at CVPR 2022 and our work on robust learning of labeling functions was accepted at Findings of ACL 2022! Congrats Krishnateja and all our coauthors!
Invited Speaker at the "Bi-level Optimization in Machine Learning" and "Submodular Optimization" sessions at INFORMS Optimization Society (IOS) 2022!
Two papers from CARAML lab are accepted at AAAI 2022! Congrats Krishnateja and Suraj!
Our work on Submodular Information Measures was accepted to Transactions of Information Theory Journal.
Outstanding Reviewer Award from NeurIPS 2021! (received the same award in 2020 as well).
Excited to be giving a tutorial at AAAI 2022 on Subset Selection in Machine Learning: Theory, Applications, and Hands-on. Stay tuned for more updates!
Three papers from CARAML lab are accepted at NeurIPS 2021! Congrats Krishna, Ping, Nathan, and Suraj!
Received an NSF Collaborative Medium Award on Submodular Information Functions with Applications to Machine Learning. Thanks, NSF!
Received gift funding from Adobe for Targeted Subset Selection! Thanks, Adobe!
Received gift funding from Google on Continuous Learning! Thanks, Google!
Together with Abir De, Ganesh Ramakrishnan, and Jeff Bilmes, I am co-organizing a workshop on Subset Selection in Machine Learning: From Theory to Applications at ICML 2021 on July 24th 2021! Workshop page: https://icml.cc/virtual/2021/workshop/8351
Will serve as an Area chair for AAAI 2022
Excited to release submodlib (Github: https://github.com/decile-team/submodlib), a submodular optimization toolkit. Credits to Vishal Kaushal for leading this effort.
Excited to release CORDS (Github: https://github.com/decile-team/cords), a PyTorch-based open-source efficient deep model training and autoML library! Credits to my student Krishnateja Killamsetty for leading this.
Excited to release DISTIL (Github: https://github.com/decile-team/distil), a PyTorch-based open-source active learning toolkit for deep learning! Credits to my students Nathan Beck and Durga Sivasubramanian for leading this.
Two papers (GRAD-MATCH and SELCON) accepted to ICML 2021!
Two papers on rule augmented learning accepted at Findings of ACL 2021 (one short and one long).
Invited Speaker at the London Symposium on Information Theory (LSIT) 2021 (Youtube Link to the Talk)
Happy to announce that we have released VISIOCITY, a dataset comprising of long videos for video summarization, and more broadly video understanding!
Our work on "A Clustering based Selection Framework for Cost Aware and Test-time Feature Elicitation" received Best Paper Honorable Mention at CODS-COMAD 2021! Congrats Srijita and Sriraam!
I will be presenting a tutorial on Combinatorial Approaches for Data, Topic and Feature Selection and Summarization at IJCAI 2020 with Ganesh Ramakrishnan (presented a similar one at ECAI 2020 earlier this year).
Our paper on Combinatorial Information Measures accepted for publication at ALT 2021, and our paper on Online Video Summarization accepted at SDM 2021!
Our paper on Data Subset Selection (GLISTER) accepted to AAAI 2021!
Invited Talk in the Special Session Deep Learning and Information Theory at SPCOM 2020 (Virtual)
Senior Program Committee for AAAI 2021
Selected among the 10% of the Reviewers for NeurIPS 2020
Our paper on Concave Aspects of Submodular Functions accepted at ISIT 2020.
Invited Speaker and Participant at the Workshop on Optimization in Machine Learning at IST Austria, May 2020
Invited Speaker at the Information Theory and Applications (ITA) workshop at San Diego, CA in February 2020
Our paper on Robust Submodular Minimization accepted at ECAI 2020!
I'm teaching (and designing) a new course at UT Dallas in Spring 2020 on Optimization in Machine Learning (Course Website). This course will cover the basics of both continuous and discrete optimization in ML. This course will be a mix of theory and practical (implementational) aspects of continuous and discrete optimization.
I've joined the CS Department of University of Texas, Dallas in Spring 2020 as an Assistant Professor.
September 2019: Visited Tata Institute of Fundamental Research for an Invited Talk
February 2019: Visited University of Texas at Dallas and University of Pittsburgh in February 2018 and gave a talk on Scalable and Practical Discrete Optimization for Big Data (see this link).
December 2018: Two papers accepted into AISTATS 2019!
October 2018: Tutorial Speaker at the 7th IEEE Winter Conference on Applications of Computer Vision (WACV) 2019 (see tutorial website. Slides are on the website)
October 2018: Three papers accepted to WACV 2019!
October 2018: Video Analytics software developed with collaborators at IIT Bombay available now at this link.
October 2018: Invited Talk at Allen Institute of AI and Google Seattle, October 2018 (Video Link)
July 2018: Released Open Source software Jensen with my collaborators John Halloran and Kai Wei
May 2017: Presented our work on Online Learning for Click Prediction at the Microsoft Machine Learning, AI and Data Science Conference
May 2017: Finalist at the LDV Vision Conference, New York
March 2017: Invited Speaker at AMS Sectional Meeting, Special Session on Geometry and Optimization in Computer Vision, Pullman, WA
March 2017: Our work on Limited Vocabulary Speech Data Subset Selection selected to Appear in Computer Speech & Language, 2017. Corpus Definitions and Baselines for SVitchboard-II and FiSVer-I datasets can be found at this link.
April 2016: Work on Minimizing Ratio of Submodular Function accepted at ICML 2016
Feb 2016: Finished my PostDoc. Will be Joining Microsoft, starting March 2016.
Two Papers accepted in NIPS 2015, Two Papers in AISTATS 2015, One Paper in ACL and INTERSPEECH 2015 and one paper in ICML 2015
Invited Speaker at the International Symposium on Mathematical Programming (ISMP), Pittsburg - July, 2015 (Session on Submodular Optimization, Link)
Invited Lecturer at the Non-convex Optimization for Machine Learning (NOML) Summer School, IIT Bombay, India, June 2015
Successfully defended in March 2015!
Selected Publications
For the complete list of publications and workshop papers, see my publications page or my research page.
Anay Majee, Ryan Sharp, Rishabh Iyer, SMILe: Leveraging Submodular Mutual Information For Robust Few-Shot Object Detection, In European Conference on Computer Vision, ECCV 2024
Anay Majee, Suraj Kothawade, Krishnateja Killiamsetty, Rishabh Iyer, SCoRe: Submodular Combinatorial Representation Learning for Real-World Class-Imbalanced Settings, In International Conference of Machine Learning, ICML 2024
HSVNS Kowndinya Renduchintala, Krishnateja Killamsetty, Sumit Bhatia, Milan Aggarwal, Ganesh Ramakrishnan, Rishabh Iyer, Balaji Krishnamurthy, INGENIOUS: Using Informative Data Subsets for Efficient Pre-Training of Large Language Models, In Findings of EMNLP 2023 (Long Paper)
Suraj N. Kothawade, Anmol Reddy Mekala, D.Chandra Sekhara SS Hetha Havya, Mayank Kothyari, Rishabh K. Iyer, Ganesh Ramakrishnan and Preethi Jyothi, DITTO: Data-efficient and Fair Targeted Subset Selection for ASR Accent Adaptation, In Association of Computational Linguists, ACL 2023 (Long Paper)
Krishnateja Killamsetty, Guttu Sai Abhishek, Aakriti Lnu, Alexandre V. Evfimievski, Lucian Popa, Ganesh Ramakrishnan, Rishabh K Iyer, AUTOMATA: Gradient Based Data Subset Selection for Compute-Efficient Hyper-parameter Tuning, In Neural Information Processing Systems, NeurIPS 2022
Athresh Karanam, Krishnateja Killamsetty, Harsha Kokel, Rishabh K Iyer, Orient: Submodular Mutual Information Measures for Data Subset Selection under Distribution Shift, In Neural Information Processing Systems, NeurIPS 2022
Suraj Kothawade, Saikat Ghosh, Sumit Shekhar, Yu Xiang, Rishabh Iyer, TALISMAN: Targeted Active Learning for Object Detection with Rare Classes and Slices using Submodular Mutual Information, In Proceedings of European Conference on Computer Vision, ECCV 2022
Changbin Li, Suraj Kothawade, Feng Chen, and Rishabh Iyer, PLATINUM: Semi-Supervised Model Agnostic Meta-Learning using Submodular Mutual Information, In International Conference of Machine Learning, ICML 2022
Rishabh Tiwari, Krishnateja Killamsetty, Rishabh Iyer, and Pradeep Shenoy, GCR: Gradient Coreset based Replay Buffer Selection for Continual Learning, In Computer Vision and Pattern Recognition, CVPR 2022
Rishabh Iyer, Ninad Khargonkar, Jeff Bilmes, Himanshu Asnani, Generalized Submodular Information Measures: Theoretical Properties, Examples, Optimization, Algorithms, and Applications, In IEEE Transactions of Information Theory, February 2022 (Conf version: Submodular Combinatorial Information Measures with Applications in Machine Learning appeared in ALT 2021)
Suraj Kothawade, Vishal Kaushal, Ganesh Ramakrishnan, Jeff Bilmes, Rishabh Iyer, PRISM: A Rich Class of Parameterized Submodular Information Measures for Guided Data Subset Selection, In 36th AAAI Conference on Artificial Intelligence, AAAI 2022 (15% Acceptance Rate)
Suraj Kothawade; Nathan Beck; Krishnateja Killamsetty; Rishabh Iyer, SIMILAR: Submodular Information Measures Based Active Learning In Realistic Scenarios, In Neural Information Processing Systems, NeurIPS 2021
Krishnateja Killamsetty, Xujiang Zhou, Feng Chen, and Rishabh Iyer, RETRIEVE: Coreset Selection for Efficient and Robust Semi-Supervised Learning, In Neural Information Processing Systems, NeurIPS 2021
Krishnateja Killamsetty, Durga Sivasubramanian, Ganesh Ramakrishnan, Abir De, Rishabh Iyer, GRAD-MATCH: A Gradient Matching Based Data Subset Selection for Efficient Deep Model Training, Proc. International Conference on Machine Learning ( ICML) 2021
Durga Sivasubramanian, Rishabh Iyer, Ganesh Ramakrishnan, and Abir De, Training Data Subset Selection for Regression with Controlled Validation Error, Proc. International Conference on Machine Learning ( ICML) 2021
Krishnateja Killamsetty, S Durga, Ganesh Ramakrishnan, and Rishabh Iyer, GLISTER: Generalization based Data Subset Selection for Efficient and Robust Learning, 35th AAAI Conference on Artificial Intelligence, AAAI 2021 (21% Acceptance Rate)
Ayush Maheshwari, Oishik Chatterjee, KrishnaTeja Killamsetty, Ganesh Ramakrishnan, and Rishabh Iyer, Data Programming using Semi-Supervision and Subset Selection, Findings of ACL, 2021 (Long Paper)
Srijita Das, Rishabh Iyer, Sriraam Natarajan, A Clustering based Selection Framework for Cost Aware and Test-time Feature Elicitation, In CODS-COMAD 2021 (Best Paper Honorable Mention, Research Track)
Rishabh Iyer and Jeff Bilmes, A Memoization Framework for Scaling Submodular Optimization to Large Scale Problems, Artificial Intelligence and Statistics (AISTATS) 2019, Naha, Okinawa, Japan
Vishal Kaushal, Rishabh Iyer, Suraj Kothiwade, Rohan Mahadev, Khoshrav Doctor, and Ganesh Ramakrishnan, Learning From Less Data: A Unified Data Subset Selection and Active Learning Framework for Computer Vision, 7th IEEE Winter Conference on Applications of Computer Vision (WACV), 2019 Hawaii, USA (Link to the Video)
Yuzong Liu, Rishabh Iyer, Katrin Kirchhoff, Jeff Bilmes, SVitchboard-II and FiSVer-I: Crafting high quality and low complexity conversational english speech corpora using submodular function optimization, Computer Speech & Language 42, 122-142, 2017 (Corpus Definitions and Baselines for SVitchboard-II and FiSVer-I datasets can be found at this link)
Wenruo Bai, Rishabh Iyer, Kai Wei, Jeff Bilmes, Algorithms for optimizing the ratio of submodular functions, In Proc. International Conference on Machine Learning ( ICML) 2016
Kai Wei, Rishabh Iyer, Shenjie Wang, Wenruo Bai, Jeff Bilmes, Mixed robust/average submodular partitioning: Fast algorithms, guarantees, and applications, In Advances of Neural Information Processing Systems (NIPS) 2015
Kai Wei, Rishabh Iyer, Jeff Bilmes, Submodularity in data subset selection and active learning, International Conference on Machine Learning (ICML) 2015
Sebastian Tschiatschek, Rishabh K Iyer, Haochen Wei, Jeff A Bilmes, Learning mixtures of submodular functions for image collection summarization, In Advances in Neural Information Processing Systems (NIPS) 2014
Rishabh Iyer and Jeff Bilmes, Submodular optimization with submodular cover and submodular knapsack constraints, In Advances Neural Information Processing Systems 2013 (Winner of the Outstanding Paper Award) Link to Video, from 56th Minute.
Rishabh Iyer, Stefanie Jegelka, Jeff Bilmes, Fast semidifferential-based submodular function optimization, International Conference on Machine Learning (ICML) 2013 (Winner of the Best Paper Award)
Rishabh Iyer, Jeff A Bilmes, The Lovász-Bregman Divergence and connections to rank aggregation, clustering, and web ranking, Uncertainty In Artificial Intelligence (UAI) 2013
Rishabh Iyer, Jeff Bilmes, Algorithms for approximate minimization of the difference between submodular functions, with applications, Uncertainty in Artificial Intelligence (UAI) 2012
Funding and Support
Our research is supported graciously by research grants from NSF, Google, Adobe, Amazon, and the UT Dallas startup fund. See our funding page for more details.