Karthikeyan Shanmugam
Email:
(firstname)(lastname)88 AT gmail DOT com
(firstname)vs AT google DOT com
I am a Research Scientist at Google Research India (Bengaluru) since April 2022. I am part of the Machine Learning Foundations and Optimization Team.
Previously, I was a Research Staff Member with the IBM Research AI, NY during the period 2017-2022 and a Herman Goldstine Postdoctoral Fellow at IBM Research, NY in the period 2016-2017. I obtained my Ph.D. in ECE from UT Austin in 2016. My advisor at UT was Alex Dimakis. I obtained my MS degree in Electrical Engineering (2010-2012) from the University of Southern California, B.Tech and M.Tech degrees in Electrical Engineering from IIT Madras in 2010.
My research interests broadly lie in Graph algorithms, Machine learning, Optimization, Coding Theory and Information Theory. Specifically in machine learning, my recent focus is on Causal Inference, Bandits/RL and Explainable AI.
[ Google Scholar] [Dissertation]
Publications
Journals (Accepted/Submitted)
J1. A Repair Framework for Scalar MDS Codes
K. Shanmugam, D.S. Papailiopoulos , A.G. Dimakis and G. Caire
IEEE JSAC Special issue on Distributed Storage, Vol:32(5), 998 -1007, 2014. [ arxiv ]
(Conference version C1 )
J2. FemtoCaching: Wireless Content Delivery through Distributed Caching Helpers
K. Shanmugam, N. Golrezaei , A.G. Dimakis, A.F. Molisch and G. Caire
IEEE Transactions on Information Theory, 8402-8413, Vol:59(12), Dec 2013. [ arxiv version ]
(Conference versions C2 and C3 )
J3.Finite Length Analysis of Caching-Aided Coded Multicasting
Karthikeyan Shanmugam, Mingyue Ji, Antonia M.Tulino, Jaime Llorca, Alexandros G. Dimakis.
IEEE Transactions on Information Theory, 2016. [ arxiv version] [IEEE version]
J4.Efficient Algorithms for Coded Multicasting in Heterogeneous Caching Networks
G. Vettigli, M. Ji, K. Shanmugam, J. Llorca and A.M. Tulino and Giuseppe Caire
Entropy 2019 [link]
J5.AI Explainability 360. An Extensible Toolkit of AI Explainability Algorithms.
V. Arya, R. K. E. Bellamy, P. Chen, A. Dhurandhar, M. Hind, S. C. Hoffman, S. Houde, Q. V. Liao, R. Luss, A. Mojsilović, S. Mourad, P. Pedemonte, R. Raghavendra, J. Richards, P. Sattigeri, K. Shanmugam, M. Singh, K. R. Varshney, D. Wei, Y. Zhang
Journal of Machine Learning Research (JMLR), 2020 [ link ]
J6. Interventional Fairness with Indirect Knowledge of Unobserved Protected Attributes
S. Galhotra, K. Shanmugam, P. Sattigeri, K. R. Varshney
Entropy 2021 [link]
J7. Causal Bandits for Linear Structural Equation Models
Burak Varici, Karthikeyan Shanmugam, Prasanna Sattigeri, Ali Tajer
Journal of Machine Learning Research, JMLR 2023 [arxiv link].
J8. A Lyapunov Theory for Finite-Sample Guarantees of Markovian Stochastic Approximation.
Zaiwei Chen, Siva Theja Maguluri, Sanjay Shakkottai, and Karthikeyan Shanmugam
Operations Research 2023. [pdf]
Conferences
Machine Learning
Fairness under Covariate Shift: Improving Fairness-Accuracy tradeoff with few Unlabeled Test Samples
Shreyas Havaldar, Jatin Chauhan, Karthikeyan Shanmugam, Jay Nandy, Aravindan Raghuveer
AAAI 2024 [arxiv]
Identifiability Guarantees for Causal Disentanglement from Soft Interventions
Jiaqi Zhang, Chandler Squires, Kristjan Greenewald, Akash Srivastava, Karthikeyan Shanmugam, Caroline Uhler
NeurIPS 2023 [arxiv]
Blocked Collaborative Bandits: Online Collaborative Filtering with Per-Item Budget Constraints
Soumyabrata Pal, Arun Sai Suggala, Karthikeyan Shanmugam, Prateek Jain
NeurIPS 2023 [arxiv]
Front-door Adjustment Beyond Markov Equivalence with Limited Graph Knowledge
Abhin Shah, Karthikeyan Shanmugam, Murat Kocaoglu
NeurIPS 2023 [arxiv]
InfoNCE Loss Provably Learns Cluster-Preserving Representations
A. Parulekar, L. Collins, K. Shanmugam , A. Mokhtari, S. Shakkottai
COLT 2023 [arxiv]
PAC Generalization via Invariant Representations
A. Parulekar, K. Shanmugam, S. Shakkottai
ICML 2023 [arxiv]
Optimal Algorithms for Latent Bandits with Cluster Structure
S. Pal, A. Suggala, K. Shanmugam, P. Jain
AISTATS 2023 [arxiv]
Is this the Right Neighbourhood? Accurate and Query Efficient Model Agnostic Explanations
A. Dhurandhar, K.N. Ramamurthy, K. Shanmugam
NeurIPS 2022
Finding Valid Adjustments under Non-ignorability with minimal DAG knowledge
A. Shah, K. Shanmugam, K. Ahuja
Intervention Target Estimation in the Presence of Latent Variables
B. Varici, K. Shanmugam, P. Sattigeri, A. Tajer
accepted to UAI 2022 [link]
Auto-Transfer: Learning to Route Transferrable Representations
K. Murugesan, V. Sadashivaiah, R. Luss, K. Shanmugam, Pin-Yu Chen, A. Dhurandhar
ICLR 2022 [arxiv]
Process Independence Testing in Proximal Graphical Event Models
D. Bhattacharjya, K. Shanmugam, T. Gao, D. Subramanian
CLear (Causal Learning and Reasoning) 2022 [link]
Causal Feature Selection for Algorithmic Fairness
S. Galhotra, K. Shanmugam, P. Sattigeri, K. R. Varshney
SIGMOD 2022 [arxiv]
Fourier Representations for Black-Box Optimization over Categorical Variables
H. Dadkhahi, J. Rios, K. Shanmugam, P. Das
AAAI 2022 [arxiv]
CoFrNets: Interpretable Neural Architecture Inspired by Continued Fractions
I. Puri, A. Dhurandhar, T. Pedapati, K. Shanmugam, D. Wei, K. R. Varshney
NeurIPS 2021 [link]
Scalable Intervention Target Estimation in Linear Models
B. Varici, K. Shanmugam, P. Sattigeri, A. Tajer
Finite-Sample Analysis of Off-Policy TD-Learning via Generalized Bellman Operators
Z. Chen, S. T. Maguluri, S. Shakkottai, K. Shanmugam
NeurIPS 2021 [arxiv]
Conditionally Independent Data Generation
K. Ahuja, P. Sattigeri, K. Shanmugam, D.Wei, K.N. Ramamurthy, M. Kocaoglu
UAI 2021. [link]
Leveraging Latent Features for Local Explanations
R. Luss, Pin-Yu Chen, A. Dhurandhar, P. Sattigeri, Y. Zhang, K. Shanmugam, Chun-Chen Tu,
KDD 2021. [arxiv]
High-Dimensional Feature Selection for Sample Efficient Treatment Effect Estimation
Kristjan Greenewald, Dmitriy Katz-Rogozhnikov, Karthik Shanmugam
AISTATS 2021 [arxiv]
Linear Regression Games: Convergence Guarantees to Approximate Out-of-Distribution Solutions
Kartik Ahuja, Karthikeyan Shanmugam, Amit Dhurandhar
Empirical or Invariant Risk Minimization? A Sample Complexity Perspective
Kartik Ahuja, Jun Wang, Amit Dhurandhar, Karthikeyan Shanmugam, Kush R. Varshney
Treatment Effect Estimation using Invariant Risk Minimization
Abhin Shah, Kartik Ahuja, Karthikeyan Shanmugam, Dennis Wei, Kush Varshney, Amit Dhurandhar
ICASSP 2021 [arxiv]
Active Structure Learning of Causal DAGs via Directed Clique Trees
Chandler Squires, Sara Magliacane, Kristjan Greenewald, Dmitriy Katz, Murat Kocaoglu and Karthikeyan Shanmugam
NeurIPS 2020 [ arxiv ] [ code ]
Learning Global Transparent Models consistent with Local Contrastive Explanations
Tejaswini Pedapati, Avinash Balakrishnan, Karthikeyan Shanmugam and Amit Dhurandhar
NeurIPS 2020 [ arxiv ]
Mix and Match: An Optimistic Tree-Search Approach for Learning Models from Mixture Distributions
Matthew Faw, Rajat Sen, Karthikeyan Shanmugam, Constantine Caramanis and Sanjay Shakkottai
NeurIPS 2020 [ arxiv ][ code ]
Causal Discovery from Soft Interventions with Unknown Targets: Characterization and Learning
Amin Jaber, Murat Kocaoglu, Karthikeyan Shanmugam and Elias Bareinboim
NeurIPS 2020 [ pdf ]
Finite-Sample Analysis of Contractive Stochastic Approximation Using Smooth Convex Envelopes
Zaiwei Chen, Siva Theja Maguluri, Sanjay Shakkottai and Karthikeyan Shanmugam
NeurIPS 2020 [ arxiv]
Hawkesian Graphical Event Models
Xiufan Yu, Karthikeyan Shanmugam, Debarun Bhattacharjya, Tian Gao, Dharmashankar Subramanian and Lingzhou Xue.
PGM 2020 [ pdf ]
Evaluation of Causal Inference Techniques for AIOps
Vijay Arya, Karthikeyan Shanmugam, Pooja Aggarwal, Qing Wang, Prateeti Mohapatra and Seema Nagar
(Research Track - Short Paper) CODS-COMAD 2020 [ link ]
Enhancing Simple Models by Exploiting What They Already Know
Amit Dhurandhar,Karthikeyan Shanmugam and Ronny Luss
ICML 2020 [ arxiv]
Invariant Risk Minimization Games
Kartik Ahuja, Karthikeyan Shanmugam, Kush Varshney, Amit Dhurandhar
A Multi-Channel Neural Graphical Event Model with Negative Evidence
Tian Gao, Dharmashankar Subramanian, Karthikeyan Shanmugam,Debarun Bhattacharjya, Nicholas Mattei
AAAI 2020 [ arxiv]
Event Driven Continuous Time Bayesian Networks
Debarun Bhattacharjya, Karthikeyan Shanmugam, Tian Gao,Nicholas Mattei, Kush R. Varshney, Dharmashankar Subramanian
AAAI 2020 [ pdf ]
Characterization and Learning of Causal Graphs with Latent Variables from Soft Interventions
Murat Kocaoglu*, Amin Jaber*, Karthikeyan Shanmugam*, Elias Bareinboim
NeurIPS 2019. [ link ] (* - Equal Contribution)
Sample Efficient Active Learning of Causal Trees
Kristjan Greenewald, Dmitriy Katz, Karthikeyan Shanmugam, Sara Magliacane, Murat Kocaoglu, Enric Boix Adsera, Guy Bresler
NeurIPS 2019. [ link ]
Differentially Private Distributed Data Summarization under Covariate Shift
Kanthi Sarpatwar*, Karthikeyan Shanmugam*, Venkata Sitaramagiridharganesh Ganapavarapu, Ashish Jagmohan, Roman Vaculin
NeurIPS 2019. [ arxiv] (* - Equal Contribution)
ABCD-Strategy: Budgeted Experimental Design for Targeted Causal Structure Discovery
R. Agrawal, C.Squires, K. Yang, K. Shanmugam and C. Uhler
AISTATS 2019. [ arxiv]
Size of Interventional Markov Equivalence Classes in random DAG models
D. Katz, K. Shanmugam, C. Squires and C. Uhler
AISTATS 2019. [ arxiv]
Confidence Scoring Using Whitebox Meta-models with Linear Classifier Probes
T. Chen, J. Navratil, V. Iyengar and K Shanmugam
AISTATS 2019. [ arxiv]
Explanations based on the Missing: Towards Contrastive Explanations with Pertinent Negatives
A. Dhurandhar, P. Chen, R. Luss, C. Tu, P. Ting, K. Shanmugam, P. Das
Neural Information Processing Systems 2018. [ arxiv]
Improving Simple Models with Confidence Profiles
A. Dhurandhar*, K. Shanmugam*, R. Luss and P. Olsen
Neural Information Processing Systems 2018. [ arxiv] (* - Equal Contribution)
Contextual Bandits with Stochastic Experts
R. Sen, K. Shanmugam and S. Shakkottai
AISTATS 2018. [arxiv]
Model-Powered Conditional Independence Test
R. Sen, A.T. Suresh, K. Shanmugam, Sanjay Shakkottai and Alex Dimakis
Neural Information Processing Systems 2017. [ arxiv][ code ]
Experimental Design for Learning Causal Graphs with Latent Variables
M. Kocaoglu, K. Shanmugam and Elias Bareinboim
Neural Information Processing Systems 2017.[Full Version ]
Identifying Best Interventions through Online Importance Sampling
Rajat Sen, K.Shanmugam, Alex Dimakis and Sanjay Shakkottai
accepted to ICML 2017. [ arxiv]
Contextual Bandits with Latent Confounders: An NMF Approach
Rajat Sen, K.Shanmugam, M. Kocaoglu Alex Dimakis and Sanjay Shakkottai
AISTATS 2017. [ arxiv]
Distributed Estimation of Graph 4-Profiles
E.R.Elenberg, K.Shanmugam, M.Borokhovich, and A.G.Dimakis.
World Wide Web Conference(WWW) 2016 [ arxiv][ code].
Learning Causal Graphs with Small Interventions
K. Shanmugam*, M. Kocaoglu*, A.G.Dimakis and S. Vishwanath
Neural Information Processing Systems, 2015. [ arxiv] ( * - equally contributing student authors )
Beyond Triangles: A Distributed Framework for Estimating 3-profiles of Large Graphs
E.R.Elenberg, K.Shanmugam, M.Borokhovich, and A.G.Dimakis
Proc. ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), 2015. [ arxiv][ code]
On the Information Theoretic Limits of Learning Ising Models
K. Shanmugam*, R. Tandon*, A. G. Dimakis, P. Ravikumar
Neural Information Processing Systems, 2014. [ arxiv ] ( * - equally contributing student authors )
Sparse Polynomial Learning and Graph Sketching
M. Kocaoglu*, K. Shanmugam*, A. G. Dimakis, A. Klivans
Neural Information Processing Systems, 2014. (Full Oral Presentation) [ arxiv] ( * - equally contributing student authors )
Information and Coding Theory
Coded Caching with Linear Subpacketization is Possible using Ruzsa-Szeméredi Graphs
K. Shanmugam, Antonia M. Tulino and Alex Dimakis
accepted to ISIT 2017 [ arxiv]
On approximating the sum-rate for multiple unicasts
K. Shanmugam, M. Asteris and A.G. Dimakis
International Symposium on Information Theory, ISIT 2015. [ arxiv ]
An Efficient Multiple-Groupcast Coded Multicasting Scheme for Finite Fractional Caching
M.Ji, K.Shanmugam, G.Vettigli, J.Llorca, A.Tulino and G.Caire
IEEE International Conference on Communications, ICC 2015. [link]
Finite Length Analysis of Caching-Aided Coded Multicasting
K. Shanmugam, M. Ji, A.M. Tulino, J. Llorca and A.G. Dimakis
in 52nd Annual Allerton Conference on Communication, Control, and Computing (Allerton), 2014 (invited) [link]
Bounding Multiple Unicasts through Index Coding and Locally Repairable Codes
K. Shanmugam and A.G. Dimakis
International Symposium on Information Theory (ISIT 2014). [ arxiv ]
Graph Theory versus Minimum Rank for Index Coding
K. Shanmugam, A.G. Dimakis and M. Langberg
International Symposium on Information Theory (ISIT 2014). [ arxiv ]
Index Coding Problem with Side Information Repositories
K. Shanmugam, A.G. Dimakis and G. Caire
51st Annual Allerton Conference on Communications, Control and Computing, Monticello, Illinois, 2013. [ extended arxiv version ]
Local Graph Coloring and Index Coding
K. Shanmugam, A.G. Dimakis and M. Langberg
International Symposium on Information Theory (ISIT 2013), Istanbul, 2013. [ extended arxiv version ]
A Repair Framework for Scalar MDS Codes (C1)
K. Shanmugam, D.S. Papailiopoulos, A.G. Dimakis and G. Caire
in 50th Annual Allerton Conference on Communication, Control, and Computing (Allerton), 2012 [ conf version ]
Wireless downloading delay under proportional fair scheduling with coupled service and requests: An approximated analysis
K. Shanmugam and G. Caire
in IEEE International Symposium on Information Theory Proceedings (ISIT), Boston, 2012 [ conf version ]
FemtoCaching: Wireless video content delivery through distributed caching helpers(C2)
N. Golrezaei, K. Shanmugam, A.G. Dimakis, A.F. Molisch and G. Caire
in Proceedings of IEEE INFOCOM, 2012 [ conf version ]
Wireless Video Content Delivery through Coded Distributed Caching(C3)
N. Golrezaei, K. Shanmugam, A.G. Dimakis, A.F. Molisch and G. Caire
in IEEE International Conference on Communications (ICC) 2012. [ conf version ]
Undergraduate stuff
Rate Gap Analysis for Rate-adaptive Antenna Selection and Beamforming Schemes
K. Shanmugam and S. Bhashyam
in the Proceedings of IEEE GLOBECOM 2010, Miami, FL, USA, Dec 2010. [ conf version ]
Enterprise Communications Platform Support for Integrated Location-Based Applications
J. Buford, Xiaotao Wu, R. Bajpai, S. Karthikeyan and V. Krishnaswamy
in The Second International Conference on Next Generation Mobile Applications, Services and Technologies, 2008. NGMAST 2008. [ conf version ]