Subhadip

Subhadip Mukherjee

PhD 🏅

Biography

Subhadip Mukherjee is an Assistant Professor in the Department of Electronics and Electrical Communication Engineering at the Indian Institute of Technology Kharagpur. His research interests lie at the intersection of signal processing, machine learning, optimization, and inverse problems, with a particular focus on medical image reconstruction and analysis. His work aims to develop robust, self-supervised, and theoretically grounded learning algorithms for imaging applications, especially in applications such as low-dose X-ray computed tomography.

Dr. Mukherjee's research spans diffusion models for inverse problems, data-driven regularization, bilevel optimization, and learning-to-optimize methods, with an emphasis on robustness to noise, distribution shifts, and limited supervision. A central theme of his work is bridging the gap between proof-of-concept machine learning models and deployable clinical systems, combining algorithmic innovation with theoretical guarantees.

Prior to joining IIT Kharagpur, he was an Assistant Professor at the University of Bath, UK, and held a postdoctoral position at the University of Cambridge, where he worked on learning-based approaches for imaging inverse problems and observed firsthand the challenges of translating advanced algorithms into real clinical workflows. He has taught undergraduate and graduate courses in signal processing, optimization, and mathematical foundations of machine learning, and actively supervises research projects at the undergraduate and postgraduate levels.

Publications

2022

  1. Quantization-aware phase retrieval
    International Journal of Wavelets, Multiresolution and Information Processing, Apr 2022

2020

  1. PhaseSense—Signal Reconstruction from Phase-Only Measurements via Quadratic Programming
    V. Kishore, S. Mukherjee, and C. S. Seelamantula
    In International Conference on Signal Processing and Communications (SPCOM), Apr 2020
  2. Signal denoising using the minimum-probability-of-error criterion
    APSIPA Transactions on Signal and Information Processing, Apr 2020

2018

  1. Phase retrieval from binary measurements
    IEEE Signal Processing Letters, Apr 2018
  2. Phasesplit: A variable splitting framework for phase retrieval
    In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Apr 2018
  3. Speech Enhancement Using the Minimum-probability-of-error Criterion
    In INTERSPEECH, Apr 2018
  4. SPCOM 18
    Binary compressive sensing and super-resolution with unknown threshold
    In International Conference on Signal Processing and Communications (SPCOM), Apr 2018
  5. SPCOM 18
    A singular value relaxation technique for learning sparsifying transforms
    In International Conference on Signal Processing and Communications (SPCOM), Apr 2018

2017

  1. Deep sparse coding using optimized linear expansion of thresholds
    arXiv preprint arXiv:1705.07290, Apr 2017
  2. Super-resolved nuclear magnetic resonance spectroscopy
    S. Mulleti , A. Singh, V. P. Brahmkhatri, K. Chandra, T. Raza , S. P. Mukherjee, and  others
    Scientific reports, Apr 2017
  3. DNNs for sparse coding and dictionary learning
    In NIPS Bayesian Deep Learning Workshop, Apr 2017
  4. Online Reweighted Least Squares Algorithm for Sparse Recovery and Application to Short-Wave Infrared Imaging
    S. Mukherjee, H. Chen, A. Veeraraghavan, and C. S. Seelamantula
    arXiv preprint arXiv:1706.09585, Apr 2017

2016

  1. ℓ₁-K-SVD: A robust dictionary learning algorithm with simultaneous update
    Signal Processing, Apr 2016
  2. ICASSP 16
    Joint dictionary training for bandwidth extension of speech signals
    In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Apr 2016
  3. NCC 16
    Convergence rate analysis of smoothed LASSO
    In Twenty second national conference on communication (NCC), Apr 2016
  4. ICASSP 16
    A divide-and-conquer dictionary learning algorithm and its performance analysis
    In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Apr 2016
  5. arXiv 16
    Super-Resolution From Binary Measurements With Unknown Threshold
    arXiv preprint arXiv:1606.03472, Apr 2016

2014

  1. IEEE TSP 14
    Fienup algorithm with sparsity constraints: Application to frequency-domain optical-coherence tomography
    IEEE Transactions on Signal Processing, Apr 2014
  2. ICASSP 14
    An optimum shrinkage estimator based on minimum-probability-of-error criterion and application to signal denoising
    In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Apr 2014
  3. arXiv 14
    A split-and-merge dictionary learning algorithm for sparse representation
    arXiv preprint arXiv:1403.4781, Apr 2014
  4. CoRR 14
    A Robust Dictionary Learning Algorithm for Image Denoising
    CoRR, Apr 2014
  5. DSP 14
    A split-and-merge dictionary learning algorithm for sparse representation: Application to image denoising
    In 19th International Conference on Digital Signal Processing (DSP), Apr 2014

2012

  1. ICASSP 12
    An iterative algorithm for phase retrieval with sparsity constraints: application to frequency domain optical coherence tomography
    In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Apr 2012
  2. STSIP 12
    A non-iterative phase retrieval algorithm for minimum-phase signals using the annihilating filter
    Sampling Theory in Signal and Image Processing, Apr 2012