Madhav Nimishakavi
Phd Student
Computer Science & Automation, IISc, Bangalore, India
I am broadly interested in Machine Learning and its applications in Natural Language Processing. The focus of my research is on developing tensor-algebraic methods for Machine Learning, and Deep Learning on graphs and hypergraphs. I play tennis and I like to travel.
Education
- Phd, CSA, Indian Institute of Science, Bangalore, 2014-Current.
Advisor: Partha Talukdar
- ME, CSA, Indian Institute of Science, Bangalore, 2010-2012.
Advisor: M. Narasimha Murty
THESIS: Use of semantics in topic based classification [ pdf]
Work Experience
- Amazon Core Machine Learning Team, Bangalore. Applied Science Intern, June 2017 - September 2017
- Indian Institute of Science, Bangalore.Teaching Assistant, UE 101, Algorithms and Programming, Fall 2016
- Wipro, Bangalore. August 2015-October 2015, Research Intern
- Indian Institute of Science, Bangalore. Teaching Assistant, E0 268, Data Mining, Spring 2015
- Microsoft India Development Center, Hyderabad. June 2012-July 2014, Software Development Engineer in Test
News
- Sep 2019: Our paper on Hypergraph Convolutional Networks is accpeted at NeurIPS 2019.
- May 2019: Invited to talk about Deep Learning methods for Graph-based SSL at Amazon, Bengaluru, India.
- January 2019: Invited to talk about Low-rank Tensor Completion at Cods-Comad 2019, Kolkata, India.
- October 2018: Received partial travel grant from Microsoft Research for attending NIPS. Thanks to MSR.
- August 2018: Our paper on Tensor Completion is accepted in NIPS.
- August 2018: Received partial travel grant from Microsoft Research for attending CIKM. Thanks to MSR.
- August 2018: Our paper on Multi-Aspect Streaming Tensor Completion is accepted in CIKM.
- May 2018: Selected as student volunteer in ACL.
- May 2018: Received partial travel grant from Microsoft Research for attending ACL. Thanks to MSR.
- April 2018: Our paper on Higher-order Relation Schema Induction is accepted in ACL.
- March 2018: Attending Google AI/ML Workshop, Bangalore.
- January 2018: Attending Academic Research Summit 2018: A Future with AI. Organized by Microsoft Research and ACM India.
- September 2017: Attending Google NLP Summit, Zurich.
- July 2017: Attending Google NLU/NLP workshop, NewYork.
- June 2017: Joined Amazon Core ML team as an Applied Science Intern.
- Dec 2016: Attending VLAI workshop, Infosys, Mysore.
- Oct 2016: Received student scholarship from EMNLP. Thanks to EMNLP.
- Oct 2016: Received travel grant from Google to participate in EMNLP. Thanks to Google.
- Jul 2016: Our paper on Relation Schema Induction using Tensor Factorization using Side Information is accepted in EMNLP.
- Jul 2016: Finished the Comprehensive Exam.
- Jan 2016: Our lab secures 10th position worldwide in Allen AI Science Challange .
- Aug 2015: Invited to Extreme Classification Workshop organozed by Microsoft Research, India.
Preprints
- Naganand Yadati, Vikram Nitin, Madhav Nimishakavi, Prateek Yadav, Anand Louis and Partha Talukdar. Link Prediction in Hypergraphs using Graph Convolutional Networks.[pdf]
- Madhav Nimishakavi , Uday Saini and Partha Talukdar. Technical report on Applicability of Tensor Factorization methods for the problem of Predicate Induction.[ pdf].
Conference Proceedings
- Naganand Yadati, Madhav Nimishakavi, Prateek Yadav, Anand Louis and Partha Talukdar. HyperGCN:A New Method of Training Graph Convolutional Networks on Hypergraphs. To appear at NeurIPS 2019, Vancouver, Canada. [arXiv:1809.02589]
- Prateek Yadav, Madhav Nimishakavi, Naganand Yadati, Arun Rajkumar and Partha Talukdar. Lovasz Convolutional Networks. The 22rd International Conference on Artificial Intelligence and Statistics (AISTATS 2019), Okinawa, Japan. [arXiv:1805.11365]
- Madhav Nimishakavi, Pratik Jawanpuria, and Bamdev Mishra. A dual framework for low rank tensor completion. Thirty-Second Conference on Neural Information Processing Systems (NIPS 2018), Montreal, Canada.[arxiv:1712.01193] [code]. A shorter version was accepted to NIPS workshop on Synergies in Geometric Data Analysis, NIPS, 2017.
- Madhav Nimishakavi, Bamdev Mishra, Manish Gupta and Partha Talukdar. Inductive Framework for Multi-Aspect Streaming Tensor Completion with Side Information. 27th International Conference on Information and Knowledge Management (CIKM 2018), Turin, Italy. [arxiv:1802.06371] [code] [ slides]
Contact
cse.madhav@gmail.com