Hi! I’m Abhishek, an Applied Scientist at Amazon International Machine Learning

Amazon International Machine Learning (IML) is a group of ~100 ML researchers working across Bangalore, Seattle, Sydney and Berlin, led by Rajeev Rastogi. I research and train models for AutoML and NLP projects.

Bio: From 2017 to 2019 I was a Software Engineer. In late-2019, I moved to Amazon IML as a Research Engineer to work on an Automated Machine Learning platform, EPS. I got hooked on ML research and by Fall 2020 had started a CS Masters at UT Austin to sharpen the saw. At the end of 2021, two things happened: (1) I published enough internal papers to become an Applied Scientist (2) I caught the NLP bug and started working on a task-oriented chatbot (training models for intent detection, information retrieval and knowledge-base curation). Before all this, I completed my undergraduate at VJTI Mumbai in 2017.

Impact: My work has been published at internal venues such as Amazon Machine Learning Conference (AMLC). The AutoML platform (where I am now a technical lead) has been used by internal teams to deploy 1,000+ models across 14 Amazon websites, driving over $550 million dollars in revenue for Amazon businesses like Subscribe and Save. My work has also been adopted in external features like AutoGluon’s infer_limit feature and presented at ARD 2021 and AIMLSys 2022.

Future Plans: In 2023, my focus will be NLP for dialogue and publishing my work externally. I’ve also taken a background interest in scaling Neural Networks; efforts in this direction include designing architecture for scheduling multiple experiments across a 800-GPU AWS cluster, which will be used to pre-train multimodal Transformers on a billion-row dataset.

Publications

Abstracts

(Preprint) Abhishek Divekar, Mudit Agarwal, Srujana Merugu, and Nikhil Rasiwasia. "Unsupervised text augmentation using Pre-trained Paraphrase Generation".

Abhishek Divekar*, Gaurav Manchanda*, Prit Raj, Abhishek Das, Karan Tanwar, Akshay Jagatap, Vinayak Puranik, Jagannathan Srinivasan, Ramakrishna Nalam, and Nikhil Rasiwasia. "Squeezing the last DRiP: AutoML for cost-constrained product classification". 9th conference of Amazon Machine Learning (AMLC 2021) (Poster) (acceptance-rate: 30%) (Talk at Amazon Research Days 2021)

Andrew Borthwick, Abhishek Divekar, Nick Erickson, Fayaz Ahmed Farooque, Oleg Kim, Nikhil Rasiwasia, and Ethan Xu. "The CPP Multimodal AutoML Corpus and Benchmark". 1st AMLC Workshop on MultiModal Learning and Fusion at the 9th conference of Amazon Machine Learning (AMLC 2021) (Oral) (Internal venue)

Abhishek Divekar, Vinayak Puranik, Zhenyu Shi, Jinmiao Fu, and Nikhil Rasiwasia. "LEAP: LEAf node Predictions in the wild". 2nd ASCS Applied Science Workshop, 2021 (Oral) (Internal venue)

Gaurav Manchanda*, Abhishek Divekar*, Akshay Jagatap, Prit Raj, Vinayak Puranik, Nikhil Rasiwasia, Ramakrishna Nalam, and Jagannathan Srinivasa. "Entity Prediction Service: a configurable, end-to-end AutoML system". Workshop on Automated Machine Learning at the 8th conference of Amazon Machine Learning (AMLC 2020) (Poster) (Internal venue)

Abhishek Divekar, Meet Parekh, Vaibhav Savla, Rudra Mishra, and Mahesh Shirole. "Benchmarking datasets for Anomaly-based Network Intrusion Detection: KDD CUP 99 alternatives". IEEE International Conference on Computing, Communication and Security (ICCCS 2018) (Oral) (https://arxiv.org/abs/1811.05372)