About

Arunava Banerjee, Ph.D., is an Associate Professor in the Department of Computer & Information Science & Engineering. His primary research areas include Machine Learning, Computational Neuroscience, Computer Vision, and Operational Research. He earned his Ph.D. from Rutgers University in 2001. His research interests focus on understanding neural computation through the lens of machine learning and computational neuroscience, contributing to the development of models and algorithms that address complex problems in these fields.

Research topics

• Computer Science
• Artificial Intelligence
• Machine Learning
• Engineering
• Simulation
• Computational biology
• Embedded system
• Electrical engineering
• Algorithm
• Genetics
• Virology
• Computer hardware
• Real-time computing
• Telecommunications
• Biology
• Acoustics

Selected publications

• Robust Online Reconstruction of Continuous-Time Signals From a Lean Spike Train Ensemble Code

  IEEE Transactions on Signal Processing · 2025-01-01

  articleSenior author

  Sensory stimuli in animals are encoded into spike trains by neurons, offering advantages such as sparsity, energy efficiency, and high temporal resolution. This paper presents a signal processing framework that deterministically encodes continuous-time signals into biologically feasible spike trains, and addresses the questions about representable signal classes and reconstruction bounds. The framework considers encoding of a signal through spike trains generated by an ensemble of neurons using a convolve-then-threshold mechanism with various convolution kernels. A closed-form solution to the inverse problem, from spike trains to signal reconstruction, is derived in the Hilbert space of shifted kernel functions, ensuring sparse representation of a generalized Finite Rate of Innovation (FRI) class of signals. Additionally, inspired by real-time processing in biological systems, an efficient iterative version of the optimal reconstruction is formulated that considers only a finite window of past spikes, ensuring robustness of the technique to ill-conditioned encoding; convergence guarantees of the windowed reconstruction to the optimal solution are then provided. Experiments on a large audio dataset demonstrate excellent reconstruction accuracy at spike rates as low as one-third of the Nyquist rate, while showing clear competitive advantage in comparison to state-of-the-art sparse coding techniques in the low spike rate regime.

  PublisherDOI

• Gaussian Recombining Split Tree

  arXiv (Cornell University) · 2024-05-25

  preprintOpen accessSenior author

  Binomial trees are widely used in the financial sector for valuing securities with early exercise characteristics, such as American stock options. However, while effective in many scenarios, pricing options with CRR binomial trees are limited. Major limitations are volatility estimation, constant volatility assumption, subjectivity in parameter choices, and impracticality of instantaneous delta hedging. This paper presents a novel tree: Gaussian Recombining Split Tree (GRST), which is recombining and does not need log-normality or normality market assumption. GRST generates a discrete probability mass function of market data distribution, which approximates a Gaussian distribution with known parameters at any chosen time interval. GRST Mixture builds upon the GRST concept while being flexible to fit a large class of market distributions and when given a 1-D time series data and moments of distributions at each time interval, fits a Gaussian mixture with the same mixture component probabilities applied at each time interval. Gaussian Recombining Split Tre Mixture comprises several GRST tied using Gaussian mixture component probabilities at the first node. Our extensive empirical analysis shows that the option prices from the GRST align closely with the market.

  PublisherOA PDFDOI

• Robust online reconstruction of continuous-time signals from a lean spike train ensemble code

  arXiv (Cornell University) · 2024-08-12

  preprintOpen accessSenior author

  Sensory stimuli in animals are encoded into spike trains by neurons, offering advantages such as sparsity, energy efficiency, and high temporal resolution. This paper presents a signal processing framework that deterministically encodes continuous-time signals into biologically feasible spike trains, and addresses the questions about representable signal classes and reconstruction bounds. The framework considers encoding of a signal through spike trains generated by an ensemble of neurons using a convolve-then-threshold mechanism with various convolution kernels. A closed-form solution to the inverse problem, from spike trains to signal reconstruction, is derived in the Hilbert space of shifted kernel functions, ensuring sparse representation of a generalized Finite Rate of Innovation (FRI) class of signals. Additionally, inspired by real-time processing in biological systems, an efficient iterative version of the optimal reconstruction is formulated that considers only a finite window of past spikes, ensuring robustness of the technique to ill-conditioned encoding; convergence guarantees of the windowed reconstruction to the optimal solution are then provided. Experiments on a large audio dataset demonstrate excellent reconstruction accuracy at spike rates as low as one-fifth of the Nyquist rate, while showing clear competitive advantage in comparison to state-of-the-art sparse coding techniques in the low spike rate regime.

  PublisherOA PDFDOI

• Beyond Rate Coding: Signal Coding and Reconstruction Using Lean Spike Trains

  2023-05-05 · 1 citations

  articleSenior author

  Recent years have seen a growing interest in spike based encoding of continuous time signals–a hallmark of biological computation. In this context, we present a mathematical framework for signal representation, leveraging a simple but robust mechanistic model of a biologically plausible spiking neuron. The framework considers encoding of a signal through spike trains generated by an ensemble of neurons via a standard convolve-then-threshold mechanism, albeit with a wide variety of convolution kernels. Reconstruction is posited as a convex optimization minimizing energy. Formal conditions under which perfect and approximate reconstruction of the signal from the spike trains is possible are then identified. The strength of the framework is shown in experiments on a large audio dataset, demonstrating good reconstruction at a spike rate of one fifth the Nyquist rate. Comparison against a benchmark sparse coding technique, viz convolutional orthogonal matching pursuit, shows competitive results in reconstruction with orders of magnitude improvement in runtime efficiency.

  PublisherDOI

• Smart Blind Stick

  Lecture notes in electrical engineering · 2022 · 3 citations

  • Computer Science
  • Real-time computing
  • Engineering
  PublisherDOI

• Signal Coding and Reconstruction using Spike Trains

  2021-05-04

  articleSenior author

  In many animal sensory pathways, the transformation from external stimuli to spike trains is essentially deterministic. In this context, a new mathematical framework for coding and reconstruction, based on a biologically plausible model of the spiking neuron, is presented. The framework considers encoding of a signal through spike trains generated by an ensemble of neurons via a standard convolve-then-threshold mechanism, albeit with a wide variety of convolution kernels. Neurons are distinguished by their convolution kernels and threshold values. Reconstruction is posited as a convex optimization minimizing energy. Formal conditions under which perfect reconstruction of the signal from the spike trains is possible are then identified. Coding experiments on a large audio dataset are presented to demonstrate the strength of the framework.

  Publisher

• Applying machine learning to predict viral assembly for adeno-associated virus capsid libraries

  Molecular Therapy — Methods & Clinical Development · 2020 · 50 citations

  • Computer Science
  • Artificial Intelligence
  • Machine Learning

  analysis using ML computer algorithms may augment smarter and more robust libraries for selection. In this study, data of AAV capsid libraries gathered before and after viral assembly are used to train ML algorithms. We found that two ML computer algorithms, artificial neural networks (ANNs), and support vector machines (SVMs), can be trained to predict whether unknown capsid variants may assemble into viable virus-like structures. Using the most accurate models constructed, hypothetical mutation patterns in library construction were simulated to suggest the importance of N495, G546, and I554 in AAV2-derived capsids. Finally, two comparative libraries were generated using ML-derived data to biologically validate these findings and demonstrate the predictive power of ML in vector design.

  PublisherOA PDFDOI

• Spike-Triggered Descent

  arXiv (Cornell University) · 2020

  Senior authorCorresponding
  • Computer Science
  • Computer Science
  • Artificial Intelligence

  The characterization of neural responses to sensory stimuli is a central problem in neuroscience. Spike-triggered average (STA), an influential technique, has been used to extract optimal linear kernels in a variety of animal subjects. However, when the model assumptions are not met, it can lead to misleading and imprecise results. We introduce a technique, called spike-triggered descent (STD), which can be used alone or in conjunction with STA to increase precision and yield success in scenarios where STA fails. STD works by simulating a model neuron that learns to reproduce the observed spike train. Learning is achieved via parameter optimization that relies on a metric induced on the space of spike trains modeled as a novel inner product space. This technique can precisely learn higher order kernels using limited data. Kernels extracted from a Locusta migratoria tympanal nerve dataset demonstrate the strength of this approach.

  PublisherOA PDF

• Representing and Transforming Sensory Stimuli using Spike Trains

  2020-05-19

  article1st authorCorresponding
  Publisher

• Circuit for Cardinal Direction 2

  Figshare · 2019-01-01

  datasetOpen access1st authorCorresponding

  Full circuit for Cardinal direction 2

  PublisherDOI

Frequent coauthors

• Anand Rangarajan

  11 shared

• Baba C. Vemuri

  University of Florida

  11 shared

• Karthik S. Gurumoorthy

  9 shared

• Haym Hirsh

  Cornell University

  8 shared

• Hailong Meng

  Yale University

  8 shared

• Subhajit Sengupta

  NorthShore University HealthSystem

  7 shared

• Ajit Rajwade

  7 shared

• Ritwik Kumar

  7 shared

Education

• PhD, Computer Science

  Rutgers University New Brunswick

  2001

Awards & honors

• UF Term Professorship, 2018