Google Scholar Profile


Privacy-preserving Social Ambiance Measure
from Free-living Speech Associates with
Chronic Depressive and Psychotic Disorders.

Wenwan Chen, Ashutosh Sabharwal, Erica Taylor, Ankit B. Patel and
Nidal Moukaddam. To appear in Frontiers in Psychiatry (2021).

Domain-driven models yield better predictions at lower cost than reservoir computers in Lorenz systems.
Ryan Pyle, Nikola Jovanovic, Devika Subramanian, Krishna V. Palem and Ankit B. Patel. Philosophical Transactions of the Royal Society A 379: 20200246.

Justin Sahs, Ryan Pyle, Aneel Damaraju, Josue Ortega Caro, Onur Tavaslioglu, Andy Lu, Ankit Patel (2020-2021). Shallow Univariate ReLu Networks as Splines: Initialization, Loss Surface, Hessian, & Gradient Flow Dynamics. ArXiV

Josue Ortega Caro, Yilong Ju, Fabio Anselmi, Sourav Dey, Ryan Pyle, Ankit Patel (2020-2021). Using Learning Dynamics to Explore the Role of Implicit Regularization in Adversarial Examples. ArXiV


Bongard-LOGO: A new benchmark for human-level concept learning and reasoning
Weili NieZhiding YuLei MaoAnkit B. PatelYuke ZhuAnima Anandkumar
Advances in Neural Information Processing Systems (NeurIPS), 2020 (Spotlight)
arXiv / code / slides

Anselmi, Fabio, Patel, Ankit B. & Rosasco, L.
Neurally plausible mechanisms for learning selective and invariant representations. J. Math. Neurosc. 10, 12 (2020).

Weili NieTero KarrasAnimesh GargShoubhik DebnathAnjul PatneyAnkit B. PatelAnima Anandkumar (2020). 
Semi-Supervised StyleGAN for Disentanglement Learning. International Conference in Machine Learning (ICML). ArXiV ICML NVIDIA Project Page

Li Yang, Zhaoqi Leng, Guangyuan Yu, Ankit Patel, Wen-Jun Hu, Han Pu (2019).
Deep Learning-Enhanced Variational Monte Carlo Method for Quantum Many-Body Physics. Physical Review Research, 2020. PRR ArXiV


Huaijin Chen, Wanjia Liu, Rhonald Lua, Rishab Goel, Yuzhong Huang, Ashok Veeraraghavan, Ankit Patel (2019). Fast Retinomorphic Event-Driven Representations for Video Gameplay and Action Recognition. IEEE Transactions in Computational Imaging, 2019. ArXiV IEEE TCI

N. Ho, T. Nguyen, A. B. Patel, A. Anandkumar, M. I. Jordan, R. G. Baraniuk. Neural Rendering Model: Joint Generation and Prediction for Semi-Supervised Learning. DeepMath 2019.

N. Ho, T. Nguyen, A. B. Patel, A. Anandkumar, M. I. Jordan, R. G. Baraniuk. The Latent-Dependent Deep Rendering Model. Workshop on Theoretical Foundations and Applications of Deep Generative Models at ICML, 2018

Li Yang, Zhaoqi Leng, Guangyuan Yu, Ankit Patel, Wen-Jun Hu, Han Pu (2019). Deep Learning-Enhanced Variational Monte Carlo Method for Quantum Many-Body Physics. ArXiV, 2019. ArXiV

Weili Nie, Ankit B. Patel (2019). Towards A Better Understanding and Regularization of GAN Training Dynamics. UAI 2019.

Joshua J. Michalenko, Ameesh Shah, Abhinav Verma, Swarat Chaudhuri, Ankit B. Patel (2019). Finite Automata Can be Linearly Decoded from Language-Recognizing RNNs. International Conference on Learning Representations (ICLR), 2019. ICLR pdf

Weili Nie, Nina Narodytska, Ankit Patel (2019). RelGAN: Relational Generative Adversarial Networks for Text Generation. International Conference on Learning Representations (ICLR), 2019. ICLR pdf

Weili Nie, Yang Zhang, Ankit Patel (2018). A Theoretical Explanation for Perplexing Behaviors of Backpropagation-based Visualizations. Proceedings of the International Conference of Machine Learning (ICML), PMLR 80:3809-3818, 2018. ICML ArXiV pdf

Huaijin Chen, Wanjia Liu, Rishab Goel, Yuzhong Huang, Ashok Veeraraghavan, Ankit Patel (2018). EDR: Retinomorphic Event-Driven Representations for Motion Vision. IEEE International Conference on Computational Photography,.


Tan Nguyen, Richard Baraniuk, Ankit Patel (2016). Semi-Supervised Learning with the Deep Rendering Mixture Model. ArXiV pdf
Ankit Patel, Tan Nguyen, Richard Baraniuk (2016). A Probabilistic Framework for Deep Learning. NIPS 2016, Barcelona, Spain. pdf NIPS
Gavin Taylor, Ryan Burmeister, Zheng Xu, Bharat Singh, Ankit Patel, Tom Goldstein (2016). Training Neural Networks without Gradients: A Scalable ADMM Approach. ICML 2016, New York City USA. ICML ArXiV
A. Patel, T. Nguyen, and R. G. Baraniuk(2016). A Probabilistic Theory of Deep Learning. Cosyne Abstracts 2016, Salt Lake City USA.


A.Mousavi, A. Patel, and R. G. Baraniuk(2015). A Deep Learning Approach to Structured Signal Recovery. 53rd Annual Allerton Conference on Communication, Control, and Computing, Sept 29-Oct 2, 2015, Allerton Park and Retreat Center, Monticello, IL, USA.
A. PatelT. Nguyen, and R. G. BaraniukA Probabilistic Theory of Deep Learning. No. 2015-1: Rice University, Department of Electrical and Computer Engineering, Mar. 15, 2015. arXiv


All research during this period was proprietary.

2003 – 2007

The Emergence of Geometric Order in Proliferating Metazoan Epithelia.  Matthew Gibson*, Ankit Patel*, Radhika Nagpal, Norbert Perrimon. Nature 42, pp. 1038-1041. Aug 31, 2006. *co-first authors
Desynchronization: A self-organizing algorithm for desynchronization and periodic resource scheduling. Ankit Patel, Julius Degesys, Radhika Nagpal. IEEE International Conference on Self-Adaptive and Self-Organizing Systems, July 2007.
DESYNC: Self-organizing Desynchronization and TDMA on Wireless Sensor NetworksJulius Degesys, Ian Rose, Ankit Patel, Radhika Nagpal. International Conference on Information Processing in Sensor Networks, April 2007.
Firefly-Inspired Sensor Network Synchronicity with Realistic Radio Effects. Geoff Werner-Allen, Geetika Tewari, Ankit Patel, Matt Welsh, Radhika Nagpal. In the ACM Conference on Embedded Networked Sensor Systems (SenSys’05), November 2005.
Determining the Optimal Time for Feature Aided Track Correlation Between Two Radars, [U]. Ankit Patel, Matthew Smith, Keh-Ping Dunn. In Conference on Missile Defense: Sensors, Environments, Architectures, November 2003.