About me

I am a Research Scientist at Meta, working on data sub-sampling algorithms for Facebook Ads. Previously, I obtained my PhD in Computer Science (Machine Learning) at UMass Amherst; my PhD research was on improving data selection efficiency with differentiable optimization layers. During my PhD, I won the NeurIPS 2020 education challenge for improving question selection efficiency in computerized adaptive tests (e.g., GRE, GMAT). During my PhD, I did two research internships at Adobe Research, where I worked on optimal bidding strategy in real-time bidding systems. Before my PhD, I was a Software Engineer (Applied Scientist) at Microsoft BingAds where I had the opportunity to work on improving click performance and ad selection algorithms. Even before that, I had amazing years at the Indian Institute of Science, Bangalore (IISc), where I worked on robust learning under noisy labels. I have been fortunate to work with some amazing researchers and publish in many exciting machine learning/data mining conferences and workshops such as AAAI, CVPR, ICML, IJCAI, KDD, and SIAM SDM. I also received the Best Student Paper Award from the IEEE Big Data (2020) conference.

Research Interest

• Meta Learning, Differentiable Optimization Layers, Bi-level Optimization
• Sequential Models, Reinforcement Learning, Inverse Reinforcement Learning
• Computational Advertising, Computer Vision, NLP, AI for Social Goods

Updates

Publications

• Learning from Sequential User Data: Models and Sample-efficient Algorithms

  Aritra Ghosh.

  Doctoral Dissertation. 2023 (UMass Amherst).

• DiFA: Differentiable Feature Acquisition

  Aritra Ghosh, and Andrew Lan.

  Proceedings of the Thirty-Seventh AAAI Conference on Artificial Intelligence. 2023 (AAAI).

  Code

• DiPS: Differentiable Policy for Sketching in Recommender Systems

  Aritra Ghosh, Saayan Mitra, and Andrew Lan.

  Proceedings of the Thirty-Sixth AAAI Conference on Artificial Intelligence. 2022 (AAAI).

  Code

• Automated scoring for reading comprehension via in-context bert tuning

  Nigel Fernandez, Aritra Ghosh, Naiming Liu, Zichao Wang, Benoît Choffin, Richard Baraniuk, and Andrew Lan.

  Proceedings of the 23rd International Conference on Artificial Intelligence in Education. 2022 (AIED).

• BOBCAT: Bilevel Optimization-Based Computerized Adaptive Testing

  Aritra Ghosh, and Andrew Lan.

  Proceedings of the 30th International Conference on Artificial Intelligence, 2021 (IJCAI).

  Code

• Contrastive Learning Improves Model Robustness Under Label Noise

  Aritra Ghosh, and Andrew Lan.

  Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2021 (CVPR WS).

  Code Video

• Option Tracing: Beyond Correctness Analysis in Knowledge Tracing

  Aritra Ghosh, Jay Raspat, and Andrew Lan.

  International conference on artificial intelligence in education, 2021 (AIED).

  Code

• Do We Really Need Gold Samples for Sample Weighting under Label Noise?

  Aritra Ghosh, and Andrew Lan.

  Proceedings of the 2021 IEEE Winter Conference on Applications of Computer Vision, 2021 (WACV).

  Code Video

• Skill-based Career Path Modeling and Recommendation

  Aritra Ghosh, Beverly Woolf, Shlomo Zilberstein, and Andrew Lan.

  Proceedings of the 2020 IEEE International Conference on Big Data (Big Data), 2020 (IEEE BigData).

  Best Student Paper Award Code

• Context-Aware Attentive Knowledge Tracing

  Aritra Ghosh, Neil Heffernan, and Andrew S Lan.

  Proceedings of the 2020 ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2020 (KDD).

  Code Video

• Optimal Bidding Strategy without Exploration in Real-time Bidding

  Aritra Ghosh, Saayan Mitra, Somdeb Sarkhel, and Viswanathan Swaminathan.

  Proceedings of the 2020 SIAM International Conference on Data Mining. Society for Industrial and Applied Mathematics, 2020 (SIAM SDM).

• Scalable Bid Landscape Forecasting in Real-time Bidding

  Aritra Ghosh, Saayan Mitra, Somdeb Sarkhel, Jason Xie, Gang Wu, and Viswanathan Swaminathan.

  Joint European Conference on Machine Learning and Knowledge Discovery in Databases. Springer, Cham, 2019 (ECML-PKDD).

• Generative Sequential Stochastic Model for Marked Point Processes

  Abhishek Sharma, Aritra Ghosh, and Madalina Fiterau.

  ICML Time Series Workshop. 2019.

• Robust Loss Functions for Deep Neural Networks

  Aritra Ghosh, Himanshu Kumar, and P. S. Sastry.

  Thirty-First AAAI Conference on Artificial Intelligence. 2017 (AAAI).

• On the Robustness of Decision Tree Learning Under Label Noise

  Aritra Ghosh, Naresh Manwani, and P. S. Sastry.

  Pacific-Asia Conference on Knowledge Discovery and Data Mining. Springer, Cham, 2017 (PAKDD).

• A Preference Approach to Reputation in Sponsored Search

  Aritra Ghosh, Dinesh Gaurav, and Rahul Agrawal.

  Proceedings of the 25th ACM International on Conference on Information and Knowledge Management. ACM, 2016 (CIKM).

• Making Risk Minimization Tolerant to Label Noise

  Aritra Ghosh, Naresh Manwani, and P. S. Sastry.

  Neurocomputing 160 (2015): 93-107.