publications

2024

1. Testing and learning structured quantum Hamiltonians

   Srinivasan Arunachalam, Arkopal Dutt, and Francisco Escudero Gutierrez

   arXiv preprint arXiv:2411.00082, 2024

   In STOC’25

   PDF
2. Polynomial-time tolerant testing stabilizer states

   Srinivasan Arunachalam, and Arkopal Dutt

   arXiv preprint arXiv:2408.06289, 2024

   In STOC’25

   PDF
3. Power of Sequential Protocols in Hidden Quantum Channel Discrimination

   Sho Sugiura, Arkopal Dutt, William J. Munro, and 2 more authors

   Phys. Rev. Lett., Jun 2024

   PDF
4. Learning Low-Degree Quantum Objects

   Srinivasan Arunachalam, Arkopal Dutt, Francisco Escudero Gutiérrez, and 1 more author

   In 51st International Colloquium on Automata, Languages, and Programming (ICALP 2024), Jun 2024

   PDF

2023

1. Optimal Algorithms for Learning Quantum Phase States

   Srinivasan Arunachalam, Sergey Bravyi, Arkopal Dutt, and 1 more author

   In 18th Conference on the Theory of Quantum Computation, Communication and Cryptography (TQC 2023), Jun 2023

   PDF
2. Active learning of quantum system Hamiltonians yields query advantage

   Arkopal Dutt, Edwin Pednault, Chai Wah Wu, and 4 more authors

   Phys. Rev. Res., Jul 2023

   PDF

2021

1. Exponential Reduction in Sample Complexity with Learning of Ising Model Dynamics

   Arkopal Dutt, Andrey Lokhov, Marc D Vuffray, and 1 more author

   In Proceedings of the 38th International Conference on Machine Learning, 18–24 jul 2021

   Abs PDF

   The usual setting for learning the structure and parameters of a graphical model assumes the availability of independent samples produced from the corresponding multivariate probability distribution. However, for many models the mixing time of the respective Markov chain can be very large and i.i.d. samples may not be obtained. We study the problem of reconstructing binary graphical models from correlated samples produced by a dynamical process, which is natural in many applications. We analyze the sample complexity of two estimators that are based on the interaction screening objective and the conditional likelihood loss. We observe that for samples coming from a dynamical process far from equilibrium, the sample complexity reduces exponentially compared to a dynamical process that mixes quickly.