Sravan Ankireddy

sravan.ankireddy@utexas.edu
sravan-ankireddy

I am a Ph.D candidate in the department of ECE at the University of Texas at Austin, advised by Hyeji Kim . I am broadly interested in problems at the intersection of Deep Learning and Information Theory. My current research focuses on using vision and language foundation model priors to build ultra-low rate image compression schemes. Additionally, I am also interested deep learning for tabular datasets and generative models for synthetic datasets.

Previously, I also worked on deep learning based channel coding schemes for PHY layer communication.

I also closely collaborate with Pramod Viswanath (Princeton), Krishna Narayanan (TAMU), Sewoong Oh (UW Seattle), Sandeep Chinchali (UT Austin).

My goal is to combine advances from multiple fields to develop low-complex solutions at the intersection of theory and systems.

In the past, I spent two wonderful years as a Research Engineer at Qualcomm working in the Physical Layer Modelling team in Bangalore, India. Previously, I graduated with Bachelors and Masters in Electrical Engineering from IIT Madras in 2019. I have been extremely fortunate to have worked with Dr. Andrew Thangaraj and Dr. Radha Krishna Ganti for my Master’s Thesis on efficient encoding and decoding of Polar Codes.

In my spare time I enjoy reading books, watching documentaries, and hiking.

news

May 13, 2024	Joined Apple as Machine Learning Intern.
May 09, 2024	Formed my Ph.D. committee and advanced to candidacy!
May 01, 2024	DeepPolar accepted to ICML 2024.
Apr 16, 2024	New Preprint! LightCode available in arXiv.
Apr 09, 2024	Nested Construction of Polar Codes via Transformers accepted to ISIT 2024.

selected publications

ICML

DeepPolar: Inventing Nonlinear Large-Kernel Polar Codes via Deep Learning

Ashwin Hebbar*, Sravan Ankireddy*, Hyeji Kim, and 2 more authors

In Proceedings of International Conference on Machine Learning (ICML) , Jul 2024

Bib PDF Code

@inproceedings{hebbar2024deeppolar,
  author = {Hebbar*, Ashwin and Ankireddy*, Sravan and Kim, Hyeji and Oh, Sewoong and Viswanath, Pramod},
  title = {DeepPolar: Inventing Nonlinear Large-Kernel Polar Codes via Deep Learning},
  booktitle = {Proceedings of International Conference on Machine Learning (ICML)},
  year = {2024},
  venue = {ICML},
  month = jul,
}

NeurIPS

Task-Aware Distributed Source Coding Under Dynamic Bandwidth

Po-han Li*, Sravan Ankireddy*, Ruihan Philip Zhao, and 5 more authors

Advances in Neural Information Processing Systems (NeurIPS), Dec 2023

Bib PDF Code

@article{li2024task,
  author = {Li*, Po-han and Ankireddy*, Sravan and Zhao, Ruihan Philip and Mahjoub, Hossein Nourkhiz and Pari, Ehsan Moradi and Topcu, Ufuk and Chinchali, Sandeep and Kim, Hyeji},
  title = {Task-Aware Distributed Source Coding Under Dynamic Bandwidth},
  journal = {Advances in Neural Information Processing Systems (NeurIPS)},
  volume = {36},
  year = {2023},
  venue = {NeurIPS},
  month = dec,
}

ISIT

Nested Construction of Polar Codes via Transformers

Sravan Ankireddy, S. Ashwin Hebbar, Heping Wan, and 2 more authors

In Proceedings of the 2024 IEEE International Symposium on Information Theory (ISIT) , Jul 2024

Bib PDF

@inproceedings{ankireddy2024nested,
  author = {Ankireddy, Sravan and Hebbar, S. Ashwin and Wan, Heping and Cho, Joonyoung and Zhang, Charlie},
  title = {Nested Construction of Polar Codes via Transformers},
  booktitle = {Proceedings of the 2024 IEEE International Symposium on Information Theory (ISIT)},
  year = {2024},
  venue = {ISIT},
  month = jul,
  pages = {1409--1414},
}

ISIT

TinyTurbo: Efficient Turbo Decoders on Edge

S. Ashwin Hebbar, Rajesh K Mishra, Sravan Ankireddy, and 3 more authors

In Proceedings of the 2022 IEEE International Symposium on Information Theory (ISIT) , Jul 2022

Bib PDF Code

@inproceedings{hebbar2022tinyturbo,
  author = {Hebbar, S. Ashwin and Mishra, Rajesh K and Ankireddy, Sravan and Makkuva, Ashok V and Kim, Hyeji and Viswanath, Pramod},
  title = {{TinyTurbo: Efficient Turbo Decoders on Edge}},
  booktitle = {Proceedings of the 2022 IEEE International Symposium on Information Theory (ISIT)},
  year = {2022},
  venue = {ISIT},
  month = jul,
  pages = {2797--2802},
}

JSAC

LightCode: Light Analytical and Neural Codes for Channels with Feedback

Sravan Ankireddy, Krishna Narayanan, and Hyeji Kim

IEEE Journal on Selected Areas in Communications (JSAC), Mar 2024

Bib PDF Code

@article{ankireddy2024lightcode,
  author = {Ankireddy, Sravan and Narayanan, Krishna and Kim, Hyeji},
  title = {LightCode: Light Analytical and Neural Codes for Channels with Feedback},
  journal = {IEEE Journal on Selected Areas in Communications (JSAC)},
  year = {2024},
  venue = {JSAC},
  month = mar,
}