Error Correction and Statistical Inference for Quantum Applications

Feb 28, 2022, 4:00 pm4:00 pm
004 Friend Center



Event Description



Quantum technologies are steadily becoming mainstream, and recent years have particularly seen several exciting developments. It is now clear that by leveraging quantum phenomena we can achieve unprecedented advances in multiple applications such as computing, communications, and networking. While small-scale realizations are possible today, a key challenge to scalability is the decoherence of quantum information. Seminal work in the 90s established that noise is not a fundamental bottleneck for quantum technologies, and today we have several strategies for performing error correction. However, bringing error correction theory to practice still poses several challenges spanning the realms of mathematics, physics,  engineering, and computer science. The design criteria for error correction also varies with applications and hardware technologies.

In this talk, I will discuss some of the challenges arising in the different applications mentioned above. In the context of quantum computing, I will describe how our algebraic methods have helped answer some important questions about logical computation that have both fundamental and practical implications. As part of this work, we have released open-source software and also addressed hardware-inspired questions in a trapped-ion system. In quantum networking, a key task is distilling entanglement between the different nodes so as to enable various network protocols such as distributed quantum computing. Since the channels between different nodes are noisy, one needs to develop reliable schemes to distribute and distill high-fidelity entanglement. I will describe our new quantum error correction based protocol that enables multipartite entanglement distillation in networks.

Statistical inference is ubiquitous in today’s world, particular settings being correcting errors in communications and data storage. Probabilistic graphical models and message passing algorithms on graphs form the engine for classical statistical inference. However, the quantum counterparts of these inference problems may need purely quantum algorithms to perform statistical inference, such as in classical communications over optical channels. Here, classical data is coded and transmitted as coherent states over the channel, and one needs to correct channel errors at the receiver through a quantum circuit that processes the received states. I will describe our work on a quantum algorithm that passes quantum messages to correct channel errors, and show that it is a promising candidate for such quantum inference tasks.

Finally, I will briefly talk about other relevant work and also share my plans for future research. Throughout the presentation, I will demonstrate how our work strengthens the link between classical and quantum error correction. I will provide the error correction and quantum background necessary for the discussion.


Narayanan Rengaswamy is a postdoctoral research associate with Prof. Bane Vasic at the University of Arizona, where he is involved in the error correction aspects of the NSF funded Center for Quantum Networks (CQN) and DoE funded Superconducting Quantum Materials and Systems (SQMS) center. He completed his Ph.D. in Electrical Engineering in May 2020 at Duke University, working under the supervision of Prof. Henry Pfister and Prof. Robert Calderbank. His dissertation ( focused on developing systematic methods to construct fault-tolerant logical operations on stabilizer quantum error correcting codes, and on optimally decoding classical codes over the quantum pure-state channel that arises in free-space optical communications. Prior to this, he completed his M.S. in Electrical Engineering in December 2015 at Texas A&M University, where he worked with Prof. Henry Pfister on cyclic polar codes. In summer 2015, he was a research intern at Alcatel-Lucent Bell Labs, Stuttgart, Germany, where he analyzed the finite-length performance of spatially-coupled LDPC codes on the binary erasure channel, under the supervision of Dr. Laurent Schmalen and Dr. Vahid Aref. His general research interests are in classical and quantum information theory, coding theory, compressed sensing and statistical inference problems. He is passionate about discovering connections between the classical and quantum information processing worlds.

More information about his work is available on his website ( and on Google Scholar (

Electrical and Computer Engineering