Research

My PhD thesis focused on the development of ML methods based on deep generative models and novel robust neural network architectures for solving inverse problems in computational imaging (e.g., MRI, CT, Diffraction Tomography, Fourier Ptychography). Now, at Distran, my research is related to the development of an audio-visual ML system for scene understanding to improve the performance and scope of their acoustic camera. Outside of work, these days, I am studying and implementing flow models (code, blog).

Google Scholar, Semantic Scholar

Publications

(* denotes equal contribution)

Journal Papers

• R. Parhi, P. Bohra, A. El Biari, M. Pourya, M. Unser, Random ReLU Neural Networks as Non-Gaussian Processes, Journal of Machine Learning Research, vol. 26, no. 19, pp. 1–31, 2025.

• S. Ducotterd, A. Goujon, P. Bohra, D. Perdios, S. Neumayer, M. Unser, Improving Lipschitz-Constrained Neural Networks by Learning Activation Functions, Journal of Machine Learning Research, vol. 25, no. 65, pp. 1–30, 2024.

• A. Goujon, S. Neumayer, P. Bohra, S. Ducotterd, M. Unser, A Neural-Network-Based Convex Regularizer for Inverse Problems, IEEE Transactions on Computational Imaging, vol. 9, pp. 781–795, 2023.

• S. Neumayer, A. Goujon, P. Bohra, M. Unser, Approximation of Lipschitz Functions Using Deep Spline Neural Networks, SIAM Journal on Mathematics of Data Science, vol. 5, no. 2, pp. 306–322, June 2023.

• P. Bohra, P. del Aguila Pla, J.-F. Giovannelli, M. Unser, A Statistical Framework to Investigate the Optimality of Signal-Reconstruction Methods, IEEE Transactions on Signal Processing, vol. 71, pp. 2043–2055, June 1, 2023.

• P. Bohra*, T.-a. Pham*, Y. Long, J. Yoo, M. Unser, Dynamic Fourier Ptychography with Deep Spatiotemporal Priors, Inverse Problems, vol. 39, no. 6, paper no. 064005, June 2023.

• P. Bohra, T.-a. Pham, J. Dong, M. Unser, Bayesian Inversion for Nonlinear Imaging Models Using Deep Generative Priors, IEEE Transactions on Computational Imaging, vol. 8, pp. 1237–1249, 2022.

• P. Bohra*, J. Campos*, H. Gupta, S. Aziznejad, M. Unser, Learning Activation Functions in Deep (Spline) Neural Networks, IEEE Open Journal of Signal Processing, vol. 1, pp. 295–309, November 19, 2020.

• P. Bohra, M. Unser, Continuous-Domain Signal Reconstruction Using Lp-Norm Regularization, IEEE Transactions on Signal Processing, vol. 68, pp. 4543–4554, August 3, 2020.

• P. Bohra*, D. Garg*, K. S. Gurumoorthy, A. Rajwade, Variance Stabilization Based Compressive Inversion under Poisson or Poisson-Gaussian Noise with Analytical Bounds, Inverse Problems, vol. 35, no. 10, October 2019.

Conference Papers

• P. Bohra, M. Unser, Computation of “Best” Interpolants in the Lp Sense, Proceedings of the Forty-Fifth IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP’20), Barcelona, Kingdom of Spain, May 4-8, 2020, pp. 5505–5509.

• P. Bohra, A. Rajwade, Poisson Low-Rank Matrix Recovery Using the Anscombe Transform, IEEE Global Conference on Signal and Information Processing (GlobalSIP), November 2018, pp. 141-145.

Workshop Papers

• P. Bohra, D. Perdios, A. Goujon, S. Emery, M. Unser, Learning Lipschitz-Controlled Activation Functions in Neural Networks for Plug-and-Play Image Reconstruction Methods, Proceedings of the Third Workshop on Deep Learning and Inverse Problems (NeurIPS’21), Virtual, December 13, 2021, pp. 1–9.

Conference Abstracts

• M. Unser, S. Ducotterd, P. Bohra, Efficient Lip-1 Spline Networks for Convergent PnP Image Reconstruction, Proceedings of the International BASP Frontiers Conference (BASP’23), Villars-sur-Ollon, Swiss Confederation, February 5-10, 2023, pp. 18.

• S. Neumayer, P. Bohra, S. Ducotterd, A. Goujon, D. Perdios, M. Unser, Analysis of 1-Lipschitz Neural Networks, Proceedings of the 2022 Oberwolfach Workshop on Mathematical Imaging and Surface Processing (OWMISP’22), Oberwolfach, Federal Republic of Germany, August 21-27, 2022, vol. 2022/38, pp. 2257–2259.

• M. Unser, P. Bohra, J. Campos, H. Gupta, S. Aziznejad, Deep Spline Neural Networks, Online Seminars on Numerical Approximation and Applications (OSNA2’20), Passau, Federal Republic of Germany, Virtual, November 9-December 3, 2020.

Talks

• Bayesian Inference for Inverse Problems: From Sparsity-Based Methods to Deep Neural Networks
  Tutorial, IEEE International Symposium on Biomedical Imaging (ISBI), Cartagena de Indias, Republic of Colombia, April 2023

• Learning Activation Functions in Neural Networks
  Machine Learning and Signal Processing (MLSP) Seminar, ENS Lyon, March 2023

• Opportunities and Challenges for Generative Adversarial Reconstruction by Distribution Matching (CryoGAN)
  SIAM Conference on Imaging Science, Virtual, March 2022

• Bayesian Image Reconstruction: From Sparsity-Based Methods to Deep Neural Networks
  Minitutorial, SIAM Conference on Imaging Science, Virtual, March 2022