Research

My research interests (right now) are broadly in all aspects of Quantum Information Theory and Quantum Computation. My current focus is on techniques that can bridge the gap between the Noisy Intermediate-Scale Quantum (NISQ) era and the Fault Tolerance era of Quantum Computing. This includes Quantum Error Correction (QEC), Quantum Error Mitigation (QEM) and ways to combine aspects of both. I am also interested in Hybrid Quantum-Classical Computation, particularly, developing techniques to make use of small scale quantum computers as helpers to classical computers, and explore what computational advantages we can gain in this regime. Although I haven’t yet worked on it, I am interested in developing High-Level Quantum Programming primitives that can aide the development of future Quantum Algorithms.

Apart from these practical interests, I also have some highly theoretical ones. Particularly… I am interested in exploring the foundations of (phenomenological) Thermodynamics, and how Information Theory interplays with the Physics of Computation. I have also dabbled in Machine Learning, Deep Learning and Neural Networks in the past (with publications in CVPR and IEEE Transactions).

I am currently running an ad-hoc research group at BRACU, titled Quantum Information Research Collaboration, (QuIRC, pronounced koo-ark). Some areas that our current work covers - Quantum Error Correction, Learning-based Quantum Error Mitigation, Shadow Tomography, Quantum Neural Networks, Quantum Channel Discrimination, Quantum Superchannels, Entanglement Purification, and Quantum Repeaters.

If you are interested in collaborating with me, give me an email!

Research Affiliations

• Visting Researcher, Theoretical Quantum Information Sciences (2023 - Present)
  Controlled Quantum Dynamics Theory Group, Blackett Laboratory, Imperial College London
  Supervisor: Myungshik Kim

• Undergraduate Researcher (2017 - 2018)
  Digital Signal Processing Research Laboratory, Bangladesh University of Engineering & Technology
  Supervisor: Md. Kamrul Hassan

Ongoing Projects

• Learning Quantum Error Mitigation using Classical Shadows, with Jishnu Mahmud and Shadman Shahriar
• Cryptographic Advantage of Quantum Secure Direct Communication, with Shihan Sajeed

Preprints

Purification and correction of quantum channels by commutation-derived quantum filters, S. Das, J. Sun, M. Hanks, B. Koczor, M. S. Kim, arXiv: 2305.00368 [quant-ph], July 2024. (Under review in Physical Review X.)

Quantumizing Classical Games: An Introduction to Quantum Game Theory, S. Das, arXiv: 2305.00368 [quant-ph].

Publications

Design of a quantum repeater using quantum circuits and benchmarking its performance on an IBM quantum computer, S. Das, M. S. Rahman, and M. Majumdar, Quantum Information Processing 20, 245 (2021).

Application of DenseNet in Camera Model Identification and Post-processing Detection, A. M. Rafi, U. Kamal, R. Hoque, A. Abrar, S. Das, R. Laganiere, and M. K. Hasan, The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops (Media Forensics), June 2019, California, USA.

Automatic Traffic Sign Detection and Recognition Using SegU-Net and a Modified Tversky Loss Function with L1-Constraint, U. Kamal, T. I. Tonmoy, S. Das and M. K. Hasan, IEEE Transactions on Intelligent Transportation Systems 21, pp. 1-13, 2019.

(Supervised) Undergraduate Dissertations

Quantum Error Correction using Quantum Convolutional Neural Networks, Niloy Deb Roy, Ridwan Shihab, Mekhala Marium, Fatema Islam Meem, Mushfiqur Rahman Heemel
We proposed a training pipeline to design Quantum Error Correction Codes for generalized noise models using Quantum Convolutional Neural Networks.

An Entropy-based Active-Learning schedule for training Neural Networks, Mehadi Hasan, Shemonto Das, Shoaib Ahmed
We proposed a training schedule based on an entropy-based uncertainty metric to reduce the number of iterations and computational cost of training neural networks.

back