I am an Applied Scientist at Amazon Web Services (AWS) where I work on building Automated Machine Learning (AutoML) on AWS Sagemaker. Before joining AWS, I spent over 2 years at A9/Amazon advertising, where I worked in the Inventory Forecasting Systems group. Before joining Amazon, I earned a PhD in Computer Science from the University of Minnesota (UMN) where I was advised by Prof. Vipin Kumar (ACM SIGKDD 2012 Innovation Award winner and author of "Introduction to Data Mining"). Prior to that, I received my Bachelors degree in Computer Science and Engineering from Indian Institute of Technology Kanpur (IITK) in 2013. At UMN, I was part of a 5-year $10M Expeditions in Computing project to understand climate change from data. My doctoral research focused on advancing traditional machine learning algorithms to use weak supervision for addressing the problem of not having adequate labeled samples for training supervised learning models. To know more about my research at UMN and elsewhere, please take a look at some of my peer-reviewed articles listed below. email | linkedin | google scholar | CV | Thesis nayak013 AT alumni DOT umn DOT edu |
Deep layers beware: Unraveling the surprising benefits of JPEG compression for image classification pre-processing
Guruprasad Nayak, Gerald Friedland
25th IEEE International Symposium on Multimedia 2023
(Acceptance Rate: 24.69%)
preprint   abstract   Published version   bibtex
Weakly Supervised Classification Using Group-Level Labels
Guruprasad Nayak, Rahul Ghosh, Xiaowei Jia, Vipin Kumar
2nd International Workshop on Data-Efficient Machine Learning (DeMaL), Knowledge Discovery and Data Mining (KDD) Conference 2021
preprint   abstract   Published version   bibtex
Semi-supervised Classification using Attention-based Regularization on Coarse-resolution Data
Guruprasad Nayak, Rahul Ghosh, Xiaowei Jia, Varun Mithal, Vipin Kumar
SIAM International Conference on Data Mining, 2020
(Acceptance Rate: 19.3%)
preprint   abstract   Published version   bibtex   Code
Spatio-temporal classification at multiple resolutions using multi-view regularization
Guruprasad Nayak, Rahul Ghosh, Xiaowei Jia, Varun Mithal, Vipin Kumar
4th International Workshop on Big Spatial Data, IEEE BigData 2019
Published version   abstract   bibtex   Code
Automated assessment of knowledge hierarchy evolution:
comparing directed acyclic graphs
Guruprasad Nayak, Sourav Dutta, Deepak Ajwani, Patrick Nicholson, Alessandra Sala
Information Retrieval Journal, 2019
preprint   Published version   abstract   bibtex   code
Classifying Heterogeneous Sequential Data by Cyclic Domain Adaptation: An Application in Land Cover Detection
Xiaowei Jia, Guruprasad Nayak, Ankush Khandelwal, Anuj Karpatne, Vipin Kumar
SIAM International Conference on Data Mining, 2019
(Acceptance rate: 22.7%)
preprint   Published version   abstract   bibtex   code
Spatial Context-Aware Networks for Mining Temporal Discriminative Period in Land Cover Detection
Xiaowei Jia, Sheng Li, Ankush Khandelwal, Guruprasad Nayak, Anuj Karpatne, Vipin Kumar
SIAM International Conference on Data Mining, 2019
(Acceptance rate: 22.7%)
preprint   Published version   abstract   bibtex   code
Automated Knowledge Hierarchy Assessment
Guruprasad Nayak, Sourav Dutta, Deepak Ajwani, Patrick Nicholson, Alessandra Sala
Second Workshop on Knowledge Graphs and Semantics for Text Retrieval, Analysis, and Understanding (KG4IR), SIGIR 2018
Published version   abstract   bibtex   code
Classifying Multivariate Time Series by Learning Sequence-level Discriminative Patterns
Guruprasad Nayak, Varun Mithal, Xiaowei Jia, Vipin Kumar
SIAM International Conference on Data Mining, 2018
(Acceptance rate: 23.2%)
preprint   Published version   abstract   bibtex   code
Mapping Burned Areas in Tropical Forests Using a Novel Machine Learning Framework
Varun Mithal*, Guruprasad Nayak*, Ankush Khandelwal, Vipin Kumar, Ramakrishna Nemani, Nikunj Oza
Remote Sensing, 2018
preprint   Published version   abstract   bibtex   code   Web-viewer
RAPT: Rare Class Prediction in Absence of True Labels
Varun Mithal, Guruprasad Nayak, Ankush Khandelwal, Vipin Kumar, Nikunj Oza, Ramakrishna Nemani
IEEE Transactions on Knowledge and Data Engineering, 2017
preprint   Published version   abstract   bibtex   code
Predict Land Covers with Transition Modeling and Incremental Learning
Xiaowei Jia, Ankush Khandelwal, Guruprasad Nayak, James Gerber, Kimberley Carlson, Paul West, Vipin Kumar
ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2017
(Acceptance rate: 17.5%)
preprint   Published version   abstract   bibtex   code   Promotional video
Predict Land Covers with Transition Modeling and Incremental Learning
Xiaowei Jia, Ankush Khandelwal, Guruprasad Nayak, James Gerber, Kimberley Carlson, Paul West, Vipin Kumar
SIAM International Conference on Data Mining, 2017
(Acceptance rate: 26.0%)
preprint   Published version   abstract   bibtex   code
Multiple Instance Learning for burned area mapping using multitemporal reflectance data
Guruprasad Nayak, Varun Mithal, Vipin Kumar
International Workshop on Climate Informatics, 2016
Published version   abstract   code