CV
Education
- Ph.D in Computer Science, Princeton University, 2013 - 2018
- B.Tech in Computer Science and Engineering, IIT Bombay, 2009 - 2013
Work experience
- 2023 - present: Principal Researcher
- Microsoft Research, Redmond
- 2020 - 2023: Senior Researcher
- Microsoft Research, Redmond
- 2018 - 2020: Postdoctoral Researcher
- Microsoft Research, Redmond
Teaching
Service and leadership
- Served on PC for ITCS 2024, FSTTCS 2023
- Co-organized the Error Correcting Codes: Theory and Practice semester program at Simons Institute, Berkeley in Spring 2024.
Honors and Awards
- Best paper award at Symposium on Theory of Computing (STOC) 2015
- Gold Medal (and ranked 9th) in the International Physics Olympiad-2009
- Institute Silver Medal given to the most outstanding student of Computer Science department, IIT Bombay
- Ranked 3rd in the national entrance test to IITs (IITJEE-2009) taken by around 400,000 students
Talks
Higher Order MDS Codes
Talk at Simons Institue, Berkeley, CA
Differentially Private Synthetic Data via Foundation Model APIs
Talk at Google Research, New York City, NY
Privacy-Preserving Machine Learning
Talk at Jovian, Bengaluru, India
Private Convex Optimization via Exponential Mechanism
Talk at Boston-area Data Privacy Seminar, Boston, MA
Fast and Memory Efficient Differentially Private-SGD via JL Projections
Talk at Google Research, Mountain View, CA
Spanoids - An Abstraction of Spanning Structures, and a Barrier for LCCs
Talk at Highlights of Algorithms (HALG) 2020,
CSPs with Global Modular Constraints
Talk at Institute for Advanced Study (IAS), Princeton, NJ
Maximally Recoverable Local Reconstruction Codes
Talk at Microsoft Research, Redmond, WA
2-Server PIR with Sub-polynomial Communication
Talk at Rutgers University, New Brunswick, NJ
Publications
On the Emergence of Thinking in LLMs I: Searching for the Right Intuition
Guanghao Ye, Khiem Pham, Xinzhi Zhang, Sivakanth Gopi, Baolin Peng, Beibin Li, Janardhan Kulkarni, Huseyin Inan, "On the Emergence of Thinking in LLMs I: Searching for the Right Intuition." arXiv preprint arXiv:2502.06773, 2025.
DiscQuant: A Quantization Method for Neural Networks Inspired by Discrepancy Theory
Jerry Chee, Arturs Backurs, Rainie Heck, Li Zhang, Janardhan Kulkarni, Thomas Rothvoss, Sivakanth Gopi, "DiscQuant: A Quantization Method for Neural Networks Inspired by Discrepancy Theory." Conference on Learning Theory (COLT), 2025.
A Geometric Perspective on the Injective Norm of Sums of Random Tensors
Afonso Bandeira, Sivakanth Gopi, Haotian Jiang, Kevin Lucca, Thomas Rothvoss, "A Geometric Perspective on the Injective Norm of Sums of Random Tensors." Symposium of Theory of Computing (STOC), 2025.
Privacy preserving user personalization using noisy ranking
Erik Anderson, Joseph PFEIFFER, Denis Charles, Aleksandr Rebrikov, John Mooring, Brandon Maslen, Davis Gilton, Sergey Yekhanin, Sivakanth Gopi, "Privacy preserving user personalization using noisy ranking." US Patent App. 18/301,943, 2024.
Selective pre-training for private fine-tuning
Da Yu, Sivakanth Gopi, Janardhan Kulkarni, Zinan Lin, Saurabh Naik, Tomasz Religa, Jian Yin, Huishuai Zhang, "Selective pre-training for private fine-tuning." Transactions on Machine Learning Research (TMLR), 2024.
Rigidity matroids and linear algebraic matroids with applications to matrix completion and tensor codes
Joshua Brakensiek, Manik Dhar, Jiyang Gao, Sivakanth Gopi, Matt Larson, "Rigidity matroids and linear algebraic matroids with applications to matrix completion and tensor codes." arXiv preprint arXiv:2405.00778, 2024.
Ranking with Multiple Objectives
Nikhil Devanur, Sivakanth Gopi, "Ranking with Multiple Objectives." arXiv preprint arXiv:2410.12139, 2024.
Privacy-preserving in-context learning with differentially private few-shot generation
Xinyu Tang, Richard Shin, Huseyin Inan, Andre Manoel, Fatemehsadat Mireshghallah, Zinan Lin, Sivakanth Gopi, Janardhan Kulkarni, Robert Sim, "Privacy-preserving in-context learning with differentially private few-shot generation." International Conference on Learning Representations (ICLR), 2024.
Improved field size bounds for higher order MDS codes
Joshua Brakensiek, Manik Dhar, Sivakanth Gopi, "Improved field size bounds for higher order MDS codes." IEEE Transactions on Information Theory, 2024.
Generalized GM-MDS: Polynomial Codes are Higher Order MDS
Joshua Brakensiek, Manik Dhar, Sivakanth Gopi, "Generalized GM-MDS: Polynomial Codes are Higher Order MDS." Proceedings of the 56th Annual ACM Symposium on Theory of Computing, 2024.
Differentially private synthetic data via foundation model apis 2: Text
Chulin Xie, Zinan Lin, Arturs Backurs, Sivakanth Gopi, Da Yu, Huseyin Inan, Harsha Nori, Haotian Jiang, Huishuai Zhang, Yin Lee, et al., "Differentially private synthetic data via foundation model apis 2: Text." International Conference on Machine Learning (ICML), 2024.
Differentially private synthetic data via foundation model apis 1: Images
Zinan Lin, Sivakanth Gopi, Janardhan Kulkarni, Harsha Nori, Sergey Yekhanin, "Differentially private synthetic data via foundation model apis 1: Images." International Conference on Learning Representations (ICLR), 2024.
AG codes achieve list decoding capacity over constant-sized fields
Joshua Brakensiek, Manik Dhar, Sivakanth Gopi, Zihan Zhang, "AG codes achieve list decoding capacity over constant-sized fields." Proceedings of the 56th Annual ACM Symposium on Theory of Computing, 2024.
Training private and efficient language models with synthetic data from llms
Da Yu, Arturs Backurs, Sivakanth Gopi, Huseyin Inan, Janardhan Kulkarni, Zinan Lin, Chulin Xie, Huishuai Zhang, Wanrong Zhang, "Training private and efficient language models with synthetic data from llms." Socially Responsible Language Modelling Research, 2023.
Textbooks are all you need
Suriya Gunasekar, Yi Zhang, Jyoti Aneja, Caio Mendes, Allie Del, Sivakanth Gopi, Mojan Javaheripi, Piero Kauffmann, Gustavo Rosa, Olli Saarikivi, et al., "Textbooks are all you need." arXiv preprint arXiv:2306.11644, 2023.
Private convex optimization in general norms
Sivakanth Gopi, Yin Lee, Daogao Liu, Ruoqi Shen, Kevin Tian, "Private convex optimization in general norms." Proceedings of the 2023 Annual ACM-SIAM Symposium on Discrete Algorithms (SODA), 2023.
Phi-2: The surprising power of small language models
Mojan Javaheripi, S{\'e}bastien Bubeck, Marah Abdin, Jyoti Aneja, Sebastien Bubeck, Caio Mendes, Weizhu Chen, Allie Del, Ronen Eldan, Sivakanth Gopi, et al., "Phi-2: The surprising power of small language models." Microsoft Research Blog, 2023.
Generic reed-solomon codes achieve list-decoding capacity
Joshua Brakensiek, Sivakanth Gopi, Visu Makam, "Generic reed-solomon codes achieve list-decoding capacity." Proceedings of the 55th Annual ACM Symposium on Theory of Computing, 2023.
Algorithmic aspects of the log-Laplace transform and a non-Euclidean proximal sampler
Sivakanth Gopi, Yin Lee, Daogao Liu, Ruoqi Shen, Kevin Tian, "Algorithmic aspects of the log-Laplace transform and a non-Euclidean proximal sampler." The Thirty Sixth Annual Conference on Learning Theory, 2023.
A construction of Maximally Recoverable LRCs for small number of local groups
Manik Dhar, Sivakanth Gopi, "A construction of Maximally Recoverable LRCs for small number of local groups." 2023 IEEE International Symposium on Information Theory (ISIT), 2023.
Flexible decoding in DNA data storage based on redundancy codes
Sergey Yekhanin, Sivakanth Gopi, "Flexible decoding in DNA data storage based on redundancy codes." US Patent 11,495,324, 2022.
Trellis based reconstruction algorithms and inner codes for dna data storage
Sergey Yekhanin, Sivakanth Gopi, Henry Pfister, Sundara Srinivasavaradhan, "Trellis based reconstruction algorithms and inner codes for dna data storage." US Patent App. 17/102,972, 2022.
Noise generation for differential privacy
Anjaneya Malpani, Jagadeesh Huliyar, Xinyun Sun, Sreeram Nivarthi, Muthukrishnan Paramasivam, Dheepak Ramaswamy, Sriradha Selvaraj, Ananthatejas Raghavan, Sergey Yekhanin, Janardhan Kulkarni, Aleksey Ashikhmin, Sivakanth Gopi, Bingran Luo, "Noise generation for differential privacy." US Patent 11,343,012, 2022.
Private convex optimization via exponential mechanism
Sivakanth Gopi, Yin Lee, Daogao Liu, "Private convex optimization via exponential mechanism." Conference on Learning Theory, 2022.
Lower bounds for maximally recoverable tensor codes and higher order MDS codes
Joshua Brakensiek, Sivakanth Gopi, Visu Makam, "Lower bounds for maximally recoverable tensor codes and higher order MDS codes." IEEE Transactions on Information Theory, 2022.
Improved maximally recoverable LRCs using skew polynomials
Sivakanth Gopi, Venkatesan Guruswami, "Improved maximally recoverable LRCs using skew polynomials." IEEE Transactions on Information Theory, 2022.
Differentially private fine-tuning of language models
Da Yu, Saurabh Naik, Arturs Backurs, Sivakanth Gopi, Huseyin Inan, Gautam Kamath, Janardhan Kulkarni, Yin Lee, Andre Manoel, Lukas Wutschitz, et al., "Differentially private fine-tuning of language models." International Conference on Learning Representations (ICLR), 2022.
Reverse concatenation of error-correcting codes in dna data storage
Sergey Yekhanin, Sivakanth Gopi, Henry Pfister, Karin Strauss, "Reverse concatenation of error-correcting codes in dna data storage." US Patent App. 16/562,183, 2021.
Trellis BMA: Coded trace reconstruction on IDS channels for DNA storage
Sundara Srinivasavaradhan, Sivakanth Gopi, Henry Pfister, Sergey Yekhanin, "Trellis BMA: Coded trace reconstruction on IDS channels for DNA storage." 2021 IEEE International Symposium on Information Theory (ISIT), 2021.
Numerical composition of differential privacy
Sivakanth Gopi, Yin Lee, Lukas Wutschitz, "Numerical composition of differential privacy." Advances in Neural Information Processing Systems, 2021.
Fast and memory efficient differentially private-sgd via jl projections
Zhiqi Bu, Sivakanth Gopi, Janardhan Kulkarni, Yin Lee, Hanwen Shen, Uthaipon Tantipongpipat, "Fast and memory efficient differentially private-sgd via jl projections." Advances in Neural Information Processing Systems, 2021.
Differentially private n-gram extraction
Kunho Kim, Sivakanth Gopi, Janardhan Kulkarni, Sergey Yekhanin, "Differentially private n-gram extraction." Advances in neural information processing systems, 2021.
Locally private hypothesis selection
Sivakanth Gopi, Gautam Kamath, Janardhan Kulkarni, Aleksandar Nikolov, Zhiwei Wu, Huanyu Zhang, "Locally private hypothesis selection." Conference on Learning Theory, 2020.
Gaussian width bounds with applications to arithmetic progressions in random settings
Jop Bri{\"e}t, Sivakanth Gopi, "Gaussian width bounds with applications to arithmetic progressions in random settings." International Mathematics Research Notices, 2020.
Differentially private set union
Sivakanth Gopi, Pankaj Gulhane, Janardhan Kulkarni, Judy Shen, Milad Shokouhi, Sergey Yekhanin, "Differentially private set union." International Conference on Machine Learning, 2020.
Spanoids—An Abstraction of Spanning Structures, and a Barrier for LCCs
Zeev Dvir, Sivakanth Gopi, Yuzhou Gu, Avi Wigderson, "Spanoids---An Abstraction of Spanning Structures, and a Barrier for LCCs." 10th Innovations in Theoretical Computer Science Conference (ITCS), 2019.
Maximally recoverable LRCs: A field size lower bound and constructions for few heavy parities
Sivakanth Gopi, Venkatesan Guruswami, Sergey Yekhanin, "Maximally recoverable LRCs: A field size lower bound and constructions for few heavy parities." Proceedings of the Thirtieth Annual ACM-SIAM Symposium on Discrete Algorithms (SODA), 2019.
CSPs with global modular constraints: algorithms and hardness via polynomial representations
Joshua Brakensiek, Sivakanth Gopi, Venkatesan Guruswami, "CSPs with global modular constraints: algorithms and hardness via polynomial representations." Proceedings of the 51st Annual ACM SIGACT Symposium on Theory of Computing, 2019.
Optimal Instance Adaptive Algorithm for the Top-$ K $ Ranking Problem
Xi Chen, Sivakanth Gopi, Jieming Mao, Jon Schneider, "Optimal Instance Adaptive Algorithm for the Top-$ K $ Ranking Problem." IEEE Transactions on Information Theory, 2018.
Locality in coding theory
Sivakanth Gopi, "Locality in coding theory." Ph.D. Thesis, 2018.
Outlaw distributions and locally decodable codes
Jop Bri{\"e}t, Zeev Dvir, Sivakanth Gopi, "Outlaw distributions and locally decodable codes." 8th Innovations in Theoretical Computer Science Conference (ITCS), 2017.
Lower bounds for 2-query LCCs over large alphabet
Arnab Bhattacharyya, Sivakanth Gopi, Avishay Tal, "Lower bounds for 2-query LCCs over large alphabet." Approximation, Randomization, and Combinatorial Optimization. (APPROX RANDOM 2017), 2017.
Locally testable and locally correctable codes approaching the Gilbert-Varshamov bound
Sivakanth Gopi, Swastik Kopparty, Rafael Oliveira, Noga Ron-Zewi, Shubhangi Saraf, "Locally testable and locally correctable codes approaching the Gilbert-Varshamov bound." Proceedings of the 2017 Annual ACM-SIAM Symposium on Discrete Algorithms (SODA), 2017.
Competitive analysis of the top-K ranking problem
Xi Chen, Sivakanth Gopi, Jieming Mao, Jon Schneider, "Competitive analysis of the top-K ranking problem." Proceedings of the Twenty-Eighth Annual ACM-SIAM Symposium on Discrete Algorithms, 2017.
Lower bounds for constant query affine-invariant LCCs and LTCs
Arnab Bhattacharyya, Sivakanth Gopi, "Lower bounds for constant query affine-invariant LCCs and LTCs." In 31st Conference on Computational Complexity (CCC), 2016.
On the number of rich lines in truly high dimensional sets
Zeev Dvir, Sivakanth Gopi, "On the number of rich lines in truly high dimensional sets." 31st International Symposium on Computational Geometry (SoCG), 2015.
2-server PIR with subpolynomial communication
Zeev Dvir, Sivakanth Gopi, "2-server PIR with subpolynomial communication." Proceedings of the forty-seventh annual ACM symposium on Theory of Computing (STOC), 2015.
Stability of Linear Threshold Functions
Sivakanth Gopi, "Stability of Linear Threshold Functions." Undergraduate Thesis, 2013.
One-bit compressed sensing: Provable support and vector recovery
Sivakant Gopi, Praneeth Netrapalli, Prateek Jain, Aditya Nori, "One-bit compressed sensing: Provable support and vector recovery." International Conference on Machine Learning, 2013.
Synthesis from incompatible specifications
Pavol {\v{C}}ern{\`y}, Sivakanth Gopi, Thomas Henzinger, Arjun Radhakrishna, Nishant Totla, "Synthesis from incompatible specifications." Proceedings of the tenth ACM international conference on Embedded software, 2012.