Akanksha Atrey
Ph.D. Candidate
Laboratory for Advanced Software Systems
University of Massachusetts Amherst
Advisor: Prashant Shenoy


About Me

My current research interests lie at the intersection of machine learning, edge computing, and privacy with a focus on applications in IoT and mobile computing. Particularly, I am interested in characterizing, evaluating and building explainable, generalizable and privacy-preserving machine learning models on resource-constrained edge environments.

My prior work in graduate school has been oriented towards employing causal modeling techniques to understand, evaluate and develop deep reinforcement learning agents.

Prior to joining UMass, I was a software engineer at IBM where I worked on the z/OS Mainframe. I completed my Bachelor of Science in Mathematics and Computer Science from the University at Albany, SUNY.


Recent News


Publications

  1. Measuring and Characterizing Generalization in Deep Reinforcement Learning.
    Sam Witty, Jun K. Lee, Emma Tosch, Akanksha Atrey, Kaleigh Clary, Michael L. Littman, David Jensen.
    Applied AI Letters 2021

  2. Preserving Privacy in Personalized Models for Distributed Mobile Services
    Akanksha Atrey, Prashant Shenoy, David Jensen
    IEEE International Conference on Distributed Computing Systems (ICDCS) 2021.

  3. Exploring E-petitioning and media: The case of #BringBackOurGirls
    Teresa M. Harrison, Catherine Dumas, Nic DePaula, Tim Fake, Will May, Akanksha Atrey, Jooyeon Lee, Lokesh Rishi, S.S. Ravi
    Government Information Quarterly 2021.

  4. New Frontiers in IoT: Networking, Systems, Reliability, and Security Challenges
    Saurabh Bagchi, Tarek F. Abdelzaher, Ramesh Govindan, Prashant Shenoy, Akanksha Atrey, Pradipta Ghosh, Ran Xu
    IEEE Internet of Things Journal 2020.

  5. Exploratory Not Explanatory: Counterfactual Analysis of Saliency Maps for Deep Reinforcement Learning
    Akanksha Atrey, Kaleigh Clary, David Jensen
    International Conference on Learning Representations (ICLR) 2020.

  6. Identifying When Effect Restoration Will Improve Estimates of Causal Effect
    Hüseyin Oktay, Akanksha Atrey, David Jensen
    SIAM International Conference on Data Mining (SDM) 2019.

  7. Measuring and Characterizing Generalization in Deep Reinforcement Learning
    Sam Witty, Jun Ki Lee, Emma Tosch, Akanksha Atrey, Michael Littman, David Jensen
    arXiv preprint arXiv:1812.02868 2018.

  8. Generalization in Deep Reinforcement Learning
    Sam Witty, Jun Ki Lee, Emma Tosch, Akanksha Atrey, Michael Littman, David Jensen
    Critiquing and Correcting Trends in Machine Learning Workshop at NeurIPS 2018.

  9. Do Diverse Social Interactions Make Us Smile More Often? Studying Smiles and Diversity Via Social Media Photos
    Vivek K. Singh, Akanksha Atrey, Saket Hegde
    ACM Multimedia (ACMMM) 2017

  10. Towards Measuring Fine-Grained Diversity Using Social Media Photographs
    Vivek K. Singh, Saket Hegde, Akanksha Atrey
    International Conference on Web and Social Media (ICWSM) 2017

  11. E-Petitioning and Online Media: The Case of# BringBackOurGirls
    Teresa M. Harrison, Catherine Dumas, Nic DePaula, Tim Fake, Will May, Akanksha Atrey, Jooyeon Lee, Lokesh Rishi, S.S. Ravi
    International Conference on Digital Government Research (dg.o) 2017 (Best paper award)

  12. E-petition Information Diffusion in Online Social Networks
    Catherine Dumas, Akanksha Atrey, Jooyeon Lee, Teresa M. Harrison, Tim Fake, Xiaoyi Zhao, S.S. Ravi
    International Conference on Digital Government Research (dg.o) 2016


Awards and Honors


Activities