WELCOME TO RISHABH IYER'S WEBPAGE

Machine Learning Researcher

 

 

 

https://lh4.googleusercontent.com/5oooS5hG9f4xOXnEw35BWt5AX4kRKNd0ZJjrwPSKQTa1m1azGxTacKEzjq4vQoGgtC_ZO_uoGOiEvyfkK7dgaRxQ-rpv9DzCrxSxOFC1rpoTwbIsKYI6=w271

ABOUT ME

I am a Research Scientist with experience in several facets of AI & Machine Learning, including:

 

·        Video Analytics

·        Deep Learning for Image Classification and Object Detection

·        Data Summarization (Video/Image/Text)

·        Active Learning and Data Subset Selection

·        Online Learning and Reinforcement Learning.

·        Classical Machine Learning and Big Data Modeling

 

 

I also have a strong theoretical understanding of Discrete and Continuous Optimization, as well as a strong experience in getting real world Machine Learning and Computer Vision problems to work! I have written several hundreds of thousands of lines of code in C++ and Python, and have worked with massive scale problems with hundreds of Millions of Data Instances I completed my Ph.D. in 2015 from University of Washington, Seattle where I worked with Jeff Bilmes. I am excited in 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 automatically learn from this.

During my Ph.D. I won best paper awards at two of the top Machine Learning Conferences, Neural Information Processing Systems (NIPS) and International Conference of Machine Learning (ICML). I also won the Microsoft Research Ph.D. Fellowship and Facebook Ph.D. Fellowship in 2014, and the Yang Award for Outstanding Graduate Student from University of Washington, Seattle. For more information, please see my Google Scholar Profile, my LinkedIn Profile, my DBLP, or my Updated Resume.

 

PUBLICATIONS

Thesis

Rishabh Iyer, Submodular Optimization and Machine Learning: Theoretical Results, Unifying and Scalable Algorithms, and Applications, Ph.D Dissertation, University of Washington, Seattle

Pre-prints

Vishal Kaushal, Sandeep Subramanium, Rishabh Iyer and Ganesh Ramakrishnan, A Framework Towards Domain Specific Video Summarization (under preparation)

Rishabh Iyer and Jeff Bilmes, Submodular Optimization and Machine learning: Unifying Algorithms and Applications (under preparation for Foundations and Trends in Machine Learning)

Rishabh Iyer, Stefanie Jegelka and Jeff Bilmes, Submodular Function Optimization: Unifying Combinatorial Algorithms and Parameters of Complexity (under preparation)

Rishabh Iyer, A Memoization Framework for Scaling Submodular Optimization to Large Scale Problems (under preparation)

Rishabh Iyer, Kai Wei, Nimit Acharya, Tanuja Bompada, Denis Charles and Eren Manavoglu, A Unified Online Learning Framework for Click Prediction (under preparation)

Rishabh Iyer, John Halloran and Kai Wei, Jensen: An Easily-Extensible C++ Toolkit for Production-Level Machine Learning and Convex Optimization, arXiv preprint arXiv:1807.06574 (GitHub Repository)

John Moore, Joel Pfeiffer, Kai Wei, Rishabh Iyer, Denis Charles, Ran Gilad-Bachrach, Levi Boyles, Eren Manavoglu, Modeling and Simultaneously Removing Bias via Adversarial Neural Networks, arXiv preprint arXiv:1804.06909

Anurag Sahoo, Vishal Kaushal, Khoshrav Doctor, Suyash Shetty, Rishabh Iyer, Ganesh Ramakrishnan, A unified multi-faceted video summarization system, arXiv preprint arXiv:1704.01466

Rishabh Iyer and Jeff Bilmes, Polyhedral aspects of submodularity, convexity and concavity, arXiv preprint arXiv:1506.07329 (Under Submission)

 

Peer Reviewed Publications

Rishabh Iyer, Nimit Acharya, Tanuja Bompada, Denis Charles and Eren Manavoglu, A Batch Online Learning Framework for Click Prediction in Search Advertisement, Microsoft Journal of Applied Research (Internal Journal), 7th Volume, Spring 2017

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

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 (Link to Video)

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

Jennifer A Gillenwater, Rishabh K Iyer, Bethany Lusch, Rahul Kidambi, Jeff A Bilmes, Submodular hamming metrics, In Advances in Neural Information Processing Systems 2015

Yuzong Liu, Rishabh Iyer, Katrin Kirchhoff, Jeff Bilmes,SVitchboard II and FiSVer I: High-Quality Limited-Complexity Corpora of Conversational English Speech, In Proc. Interspeech, 2015

Ramkrishna Bairi, Rishabh Iyer, Ganesh Ramakrishnan, Jeff Bilmes, Summarization of Multi-Document Topic Hierarchies using Submodular Mixtures, In Association of Computational Linguists (ACL) 2015

Rishabh Iyer and Jeff Bilmes, Submodular point processes with applications to machine learning, Proc. Artificial Intelligence and Statistics (AISTATS) 2015

Yoshinobu Kawahara, Rishabh Iyer, Jeffrey Bilmes, On Approximate Non-submodular Minimization via Tree-Structured Supermodularity, Artificial Intelligence and Statistics (AISTATS) 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

Kai Wei, Rishabh K. Iyer, Jeff A. Bilmes, Fast multi-stage submodular maximization, International Conference on Machine Learning (ICML-14, Link to Video)

Rishabh Iyer, Stefanie Jegelka, Jeff Bilmes, Monotone Closure of Relaxed Constraints in Submodular Optimization: Connections Between Minimization and Maximization, Uncertainty in Artificial Intelligence (UAI) 2014

Rishabh Iyer, Rushikesh Borse, Subhasis Chaudhuri, Embedding capacity estimation of reversible watermarking schemes, Sadhana 39 (6), 1357-1385

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)

Rishabh K Iyer, Stefanie Jegelka, Jeff A Bilmes, Curvature and optimal algorithms for learning and minimizing submodular functions, In Advances of Neural Information Processing Systems 2013

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) Link to Video

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 A Bilmes, Submodular-Bregman and the Lovász-Bregman divergences with applications, In Advances in Neural Information Processing Systems 2012

Rishabh Iyer, Jeff Bilmes, Algorithms for approximate minimization of the difference between submodular functions, with applications, Uncertainty in Artificial Intelligence (UAI) 2012

Ronak Shah, Rishabh Iyer, Subhasis Chaudhuri, Object mining for large video data, British Machine Vision Conference (BMVC) 2011

Workshops

Kai Wei, Rishabh Iyer, Shenjie Wang, Wenruo Bai, Jeff Bilmes, How to intelligently distribute training data to multiple compute nodes: Distributed machine learning via submodular partitioning, Neural Information Processing Society (NIPS) Workshop, Montreal, Canada 2015

Rishabh Iyer, Jeff Bilmes, Near Optimal algorithms for constrained submodular programs with discounted cooperative costs, NIPS Workshop on Discrete Optimization in Machine Learning (DISCML) 2014

Rishabh Iyer, Stefanie Jegelka, Jeff Bilmes, Mirror descent like algorithms for submodular optimization, NIPS Workshop on Discrete Optimization in Machine Learning (DISCML) 2012

Rishabh Iyer and Torsten Moller, A spatial domain optimization of sampling point-set, MITACS Globalink Research Symposium 2010.

 

AWARDS AND RECOGNITION

 

·        Selected as a 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

·        Invited for Talks at the AMS Sectional Meeting and the International Symposium for Mathematical Programming

 

WORK EXPERIENCE AND EDUCATION

 

·        March 2016 - Present, Machine Learning Scientist, Microsoft

·        March 2015 - March 2016, Post-Doctoral Researcher, University of Washington

·        September 2011 - March 2015, Ph.D Candidate, University of Washington, Seattle

·        August 2011 - May 2011, B.Tech, IIT-Bombay

 

 

CODE AND SOFTWARE

 

SMTK: A Submodular Optimization Toolkit in C++

·        Joint work with Jeff Bilmes, Kai Wei, Yuzong Liu and several others (currently maintained by Melodi Lab, University of Washington)

·        Provided the first general purpose C++ toolkit for large scale submodular function optimization, which includes a large class of algorithms and commonly used submodular functions.

·        Has several memoization and implementation tricks to speed up the algorithms (including the implementations of the Lazy Greedy, Lazier than Lazy Greedy etc.)

·        Algorithms scale to massive datasets involving ground set sizes of several million instances.

·        Enables creating applications for several summarization (document/image/video) and data selection applications in a few lines of code!

 

Jensen: An Easily-Extensible C++ Toolkit for Production-Level Machine Learning and Convex Optimization (GitHub Repository)

·        Joint work with John Halloran

·        A modular framework for Convex optimization including several common convex functions and algorithms used in Machine Learning

·        Implements several convex functions like Logistic Loss, Hinge Loss etc. and most convex optimization algorithms including LBFGS, Trust Region Newton, LBFGS-Owl, Stochastic Gradient Descent, Nesterov’s optimal algorithm, Gradient Descent with various update rules, Conjugate gradient descent etc.

·        Implements several basic Machine Learning classifiers such as L1/L2 regularized Logistic Regression, SVMs, Probit Regression etc.

 

aiSaac: A Scalable C++ deep video analytics engine

·        Implements a scalable real time and post-mortem video analytics engine with several functionalities including object detection, face detection and recognition, human detection and human subattribute recognition, vehicle detection and vehicle subattribute recognition and face age/gender recognition

·        Integrates several open source software including OpenCV, Caffe, DarkNet, DLib and LibCCV, all in a single engine!

·        Ability to train customized object detection models and image classification models

·        Enables model finetuning and transfer learning

·        Supports live streams from surveillance cameras and several video file formats

·        Enables creating video analytics applications with a few lines of code!

 

PROFESSIONAL ACTIVITY

 

·        Reviewer for Journal of Machine Learning Research (JMLR)

·        Reviewer for Journal of Discrete Applied Mathematics (DAM)

·        Reviewer for Pattern Analysis and Machine Intelligence (PAMI)

·        Reviewer for International Conference of Machine Learning (ICML) 2013 - 2018

·        Reviewer for Neural Information Processing Systems (NIPS) 2013 – 2018

·        Reviewer for Symposium of Discrete Algorithms, SODA 2019

·        Program Committee Member for Uncertainty in Artificial Intelligence (UAI) 2013 – 2016

·        Program Committee Member for American Association of Artificial Intelligence (AAAI) 2016 - 2017

·        Program Committee Member for Artificial Intelligence and Statistics (AISTATS) 2016 - 2017

 

TALKS, presentations and tutorials

 

·        LDV Vision Conference, New York 2017

·        AMS Sectional Meeting, Special Session on Geometry and Optimization in Computer Vision, Pullman, WA, March 2017

·        Google Seattle, September 2015

·        International Symposium on Mathematical Programming (ISMP), Pittsburg - July, 2015 (Session on Submodular Optimization)

·        Invited Tutorial at the Non Convex Optimization in Machine Learning, IIT Bombay, 2015

·        UW-MSR Joint Machine Learning Symposium, Redmond (invited spotlight) - Feb. 2015. (Link to Video)

·        General Electric (GE), August, 2014.

·        University of Washington Yahoo! Machine Learning Lunch - May, 2014

·        University of Washington, Trends in Optimization (TOPS) seminar - May, 2014

·        Microsoft Research, Bangalore, January, 2014.

·        Indian Institute of Science (IISc), January, 2014 (Link to Video)

·        Indian Institute of Technology, Bombay (IIT-B), March, 2013 and Feb. 2014.

·        Indian Institute of Technology, Gandhinagar (IIT-GN), Feb. 2014.

·        Neural Information Processing Systems (NIPS) – 2013 (Link to Video, from 56th Minute)

·        International Conference on Machine Learning (ICML) – 2013 (Link to Video)

·        Uncertainty in Artificial Intelligence (UAI) - 2013.

·        Uncertainty in Artificial Intelligence (UAI) - 2012.

·        Workshop on Discrete Optimization in Machine Learning (DISCML) - 2012.

·        MITACS Globalink Research Symposium - 2010.