Research output
Research & Talks.
Papers, patents, and public talks — spanning ML research, computer vision, molecular ML, and applied AI.
Publications
-
S. Jogani, A. Pol, M. Prajapati, A. Samal, K. Bhatia, J. Parmar, U. Patel, F. Shah, N. Vyas, S. Gupta. scaLR: a low-resource deep neural network-based platform for single-cell analysis and biomarker discovery. Briefings in Bioinformatics, Volume 26, Issue 3, May 2025, bbaf243. Paper
-
R. Javia, F. Shah, S. Dave. PoseSync: Robust pose-based video synchronization. arXiv:2308.12600. arXiv
-
T. Gadhiya, F. Shah, N. Vyas, V. Gharakhanyan, J. Yang, A. Holiday. Directional Variational Transformers for continuous molecular embedding. Critical Assessment of Molecular Machine Learning workshop. Paper
-
P. Soni, F. Shah, N. Vyas. Faster object tracking pipeline for real-time tracking. arXiv:2011.03910. arXiv
-
F. Shah, T. Naik, N. Vyas. LSTM-Based Music Generation. iCMLDE (IEEE), Taiwan, 2019. IEEE
-
F. Shah, B. Vikani, N. Vyas. TF_cnnvis: Visualizing Convolutional Neural Networks. NVIDIA GTC EU, Munich, 2018. Paper
-
F. Shah, P. Shah, R. Dubey. Specularity Removal for Robust Road Detection. IEEE TENCON, Singapore, 2016. IEEE
Patents
All filed with X Development LLC.
-
Depolymerization optimization platform. T. Gadhiya, F. Shah, N. Vyas, V. Gharakhanyan, J. Yang, A. Holiday. US Patent Application 17/967,723, 2023. Patent
-
Molecular structure transformers for property prediction. T. Gadhiya, F. Shah, N. Vyas, J. Yang, V. Gharakhanyan, A. Holiday. US Patent Application 17/967,685, 2023. Patent
-
Varying embedding(s) and/or action model(s) utilized in automatic generation of action set responsive to natural language request. D. Andre, R. Singh, R. Radkoff, Y. A. Madan, N. Vyas, J. Parmar, F. Shah, S. Trivedi. US Patent Application 18/142,472, 2023. Patent
-
Machine learning platform for finding solid catalysts for depolymerization reactions. A. Holiday, V. Gharakhanyan, F. Shah, N. Vyas, T. Gadhiya. US Patent Application 18/435,957, 2023. Patent
-
Generating interfacing source code. D. Andre, N. Vyas, S. Pradhan, R. Radkoff, R. Butterfoss, F. Shah, J. Parmar. US Patent 12/217,029. Patent
-
Training and application of bottleneck models and embeddings. R. Singh, D. Andre, G. R. Honke, F. Shah, N. Vyas, J. Parmar, B. M. Rosen, S. Trivedi. US Patent Application 18/224,889, 2025. Patent
Talks & Workshops
- Applications of GenAI in design, environment & finance Ahmedabad University Invited
- Retrieval-Augmented Generation ICDAR, Athens Tutorial
- Applications of LangChain Nanyang Technological University, Singapore Invited
- Prototyping LLMs ft. LangChain CIIE, IIM Ahmedabad
- NLP Canvas ICDAR, Lausanne
- Machine Learning Summer School AERFAI, Spain
- Hands-on Deep Learning Faculty Development Program, SGSITS Indore
- Python for ML & Data Science NIT Jaipur
- LSTM Playground MIND Conference, NIT Kurukshetra
- IEEE Machine Learning Summer School DA-IICT, Gandhinagar
- Demystifying optimizers used in deep learning SciPy India Conference, IIT Bombay