Aditya Kusupati

CV | Scholar | Github
Twitter | Calendar

I am a CS PhD student at University of Washington jointly advised by Ali Farhadi and Sham Kakade. My broad research interests at the moment lie in the intersection of Machine Learning, Computer Vision and Robotics (Multimodal Perception, shh! it is a secret).

Before coming here, I spent two amazing years as a Research Fellow at Microsoft Research India with Manik Varma and Prateek Jain working on "The Extremes of Machine Learning". I have a Bachelors in CS with Honours and a Minor in EE from IIT Bombay where I had the pleasure of working with Soumen Chakrabarti on geometric embeddings for Entity Typing.

  • Our BuildSys '19 paper won the Best Paper Runner-Up Award!!
  • Paper on extreme multi-label regression accepted at WSDM '20!
  • Best reviewer (top 400) award for NeurIPS '19!
  • EdgeML revamped for the next release. Tensorflow and PyTorch support inducted including CUDA optimized FastGRNN!
  • Accepted CS PhD offer at University of Washington. Time to Se(a)ttle (Settle in Seattle. I am sorry.)!

[NEW] Extreme Regression for Dynamic Search Advertising
Yashoteja Prabhu, Aditya Kusupati, Nilesh Gupta and Manik Varma
International Conference on Web Search and Data Mining (WSDM), 2020

abstract / bibtex / pdf / reviews / code / XML Repository

This paper introduces a new learning paradigm called eXtreme Regression (XR) whose objective is to accurately predict the numerical degrees of relevance of an extremely large number of labels to a data point. XR can provide elegant solutions to many large-scale ranking and recommendation applications including Dynamic Search Advertising (DSA). XR can learn more accurate models than the recently popular extreme classifiers which incorrectly assume strictly binary-valued label relevances. Traditional regression metrics which sum the errors over all the labels are unsuitable for XR problems since they could give extremely loose bounds for the label ranking quality. Also, the existing regression algorithms won't efficiently scale to millions of labels. This paper addresses these limitations through: (1) new evaluation metrics for XR which sum only the k largest regression errors; (2) a new algorithm called XReg which decomposes XR task into a hierarchy of much smaller regression problems thus leading to highly efficient training and prediction. This paper also introduces a (3) new labelwise prediction algorithm in XReg useful for DSA and other recommendation tasks.
Experiments on benchmark datasets demonstrated that XReg can outperform the state-of-the-art extreme classifiers as well as large-scale regressors and rankers by up to 50% reduction in the new XR error metric, and up to 2% and 2.4% improvements in terms of the propensity-scored precision metric used in extreme classification and the click-through rate metric used in DSA respectively. Deployment of XReg on DSA in Bing resulted in a relative gain of 58% in revenue and 27% in query coverage. XReg's source code can be downloaded from

  author    = "Prabhu, Y. and Kusupati, A. and 
  Gupta, N. and Varma, M.",
  title     = "Extreme Regression for Dynamic 
  Search Advertising",
  booktitle = "Proceedings of the ACM International 
  Conference on Web Search and Data Mining",
  month     = "February",
  year      = "2020",

One Size Does Not Fit All: Multi-Scale, Cascaded RNNs for Radar Classification
Dhrubojyoti Roy*, Sangeeta Srivastava*, Aditya Kusupati, Pranshu Jain, Manik Varma and Anish Arora
International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation (BuildSys), 2019

Oral and demo 🏆 Best Paper Runner-Up Award
abstract / bibtex / pdf / reviews / arXiv / code / poster / dataset / demo

Edge sensing with micro-power pulse-Doppler radars is an emergent domain in monitoring and surveillance with several smart city applications. Existing solutions for the clutter versus multi-source radar classification task are limited in terms of either accuracy or efficiency, and in some cases, struggle with a trade-off between false alarms and recall of sources. We find that this problem can be resolved by learning the classifier across multiple time-scales. We propose a multi-scale, cascaded recurrent neural network architecture, MSC-RNN, comprised of an efficient multi-instance learning (MIL) Recurrent Neural Network (RNN) for clutter discrimination at a lower tier, and a more complex RNN classifier for source classification at the upper tier. By controlling the invocation of the upper RNN with the help of the lower tier conditionally, MSC-RNN achieves an overall accuracy of 0.972. Our approach holistically improves the accuracy and per-class recalls over machine learning models suitable for radar inferencing. Notably, we outperform cross-domain handcrafted feature engineering with purely time-domain deep feature learning, while also being up to ~3x more efficient than a competitive solution.

  author    = "Roy*, D. and Srivastava*, S. and 
  Kusupati, A. and Jain, P. and 
  Varma, M. and Arora, A.",
  title     = "One Size Does Not Fit All: 
  Multi-Scale, Cascaded RNNs for 
  Radar Classification",
  booktitle = "Proceedings of the ACM International 
  Conference on Systems for Energy-Efficient 
  Buildings, Cities, and Transportation",
  month     = "November",
  year      = "2019",

FastGRNN: A Fast, Accurate, Stable and Tiny Kilobyte Sized Gated Recurrent Neural Network
Aditya Kusupati, Manish Singh, Kush Bhatia, Ashish Kumar, Prateek Jain and Manik Varma
Neural Information Processing Systems (NeurIPS), 2018

abstract / bibtex / pdf / reviews / arXiv / code / video / poster / blog

This paper develops the FastRNN and FastGRNN algorithms to address the twin RNN limitations of inaccurate training and inefficient prediction. Previous approaches have improved accuracy at the expense of prediction costs making them infeasible for resource-constrained and real-time applications. Unitary RNNs have increased accuracy somewhat by restricting the range of the state transition matrix's singular values but have also increased the model size as they require a larger number of hidden units to make up for the loss in expressive power. Gated RNNs have obtained state-of-the-art accuracies by adding extra parameters thereby resulting in even larger models. FastRNN addresses these limitations by adding a residual connection that does not constrain the range of the singular values explicitly and has only two extra scalar parameters. FastGRNN then extends the residual connection to a gate by reusing the RNN matrices to match state-of-the-art gated RNN accuracies but with a 2-4x smaller model. Enforcing FastGRNN's matrices to be low-rank, sparse and quantized resulted in accurate models that could be up to 35x smaller than leading gated and unitary RNNs. This allowed FastGRNN to accurately recognize the "Hey Cortana" wakeword with a 1 KB model and to be deployed on severely resource-constrained IoT microcontrollers too tiny to store other RNN models. FastGRNN's code is available at

  author    = "Kusupati, A. and Singh, M. and 
  Bhatia, K. and Kumar, A. and Jain, P. and Varma, M.",
  title     = "{FastGRNN}: A Fast, Accurate, 
  Stable and Tiny Kilobyte Sized 
  Gated Recurrent Neural Network.",
  booktitle = "Advances in 
  Neural Information Processing Systems",
  month     = "December",
  year      = "2018",

geometric embeddings

Efficient Spatial Representation for Entity-Typing
Aditya Kusupati*, Anand Dhoot* and Soumen Chakrabarti
Undergraduate Thesis, CSE IIT Bombay, 2016-17

abstract / bibtex / pdf

The project aims at creating a efficient spatial embeddings for entities and types which would be useful for various downstream tasks such as Knowledge Base Completion, Fine-Type Tagging and Question Answering.

     author = "Kusupati*, A. and Dhoot*, A. 
     and Chakrabarti, S.",
     title = "Efficient Spatial Representation 
     for Entity-Typing",
     booktitle = "Undergraduate Thesis, CSE IIT Bombay",
     year = "2016-17"


EdgeML: Machine Learning for resource-constrained edge devices
Work of many amazing collaborators. I was one of the initial and primary contributers.
Github, Microsoft Research India, 2017-present.

abstract / bibtex

Open source repository for all the research outputs on resource efficient Machine Learning from Microsoft Research India. It contains scalable and multi-framework compatible implementations of Bonsai, ProtoNN, FastCells, EMI-RNN, S-RNN, a tool named SeeDot for fixed-point compilation of ML models along with applications such as on-device Keyword spotting and Gesturepod.
EdgeML is under MIT license and is open to contributions and suggestions. Please cite the software if you happen to use EdgeML in your research or otherwise (use the latest bibtex from the repository in case this gets outdated)

    author = {{Dennis, D.~K. and Gaurkar, Y. and Gopinath, S. and Gupta, C. and
               Jain, M. and Kumar, A. and Kusupati, A. and Lovett, C. 
               and Patil, S.~G. and Simhadri, H.~V.}},
    title = {{EdgeML: Machine Learning for resource-constrained edge devices}},
    url = {},
    version = {0.3},

CS226/254: Digital Logic Design + Lab - Spring '17, IIT Bombay
Instructor: Prof. Supratik Chakraborty

CS251: Software Systems Lab - Fall '16, IIT Bombay
Instructor: Prof. Sharat Chandran

CS226/254: Digital Logic Design + Lab - Spring '16, IIT Bombay
Instructor: Prof. Supratik Chakraborty and Prof. Ashwin Gumaste

CS101: Computer Programming and Utilisation - Fall '15, IIT Bombay
Instructor: Prof. Varsha Apte

CS101: Computer Programming and Utilisation - Spring '15, IIT Bombay
Instructor: Prof. Kavi Arya

  • The Edge of Machine Learning
    • University of Washington Sensor Systems Seminar (October '19)
    • University of Washington CSE Colloquium (October '19)
    • VGG @ Oxford University, UK on (April '19)
    • Microsoft Research Redmond (March '19)
    • Microsoft Research India (August '18)
  • The Extremes of Machine Learning
    • Microsoft Bing Bellevue (March '19)

Template: this, this and this