Research:
I am keenly interested in working on the challenging topics in Trustworthy AI, especially security and privacy in machine learning.
I have also gotten some research experience exploring some topics in Digital Image Forensics, Applied Cryptography, Computer Vision & Deep learning.
Our sponsors:
Research Publications:
Javad Rafiei Asl, Prajwal Panzade, Eduardo Blanco, Daniel Takabi, Zhipeng Cai. RobustSentEmbed: Robust Sentence Embeddings Using Adversarial Self-Supervised Contrastive Learning. Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL Findings) 2024. (Accepted).
Golnoush Haddadian, Prajwal Panzade, Daniel Takabi, Min Kim. Evaluating private artificial intelligence (AI) curriculum in computer science (CS) education: Insights for advancing student-centered CS learning. The 18th International Conference of the Learning Sciences/Computer-Supported Collaborative Learning (ICLS/CSCL-2024) (Accepted).
Prajwal Panzade, Daniel Takabi and Zhipeng Cai. I Can’t See It But I Can Fine-tune It: On Encrypted Fine-tuning of Transformers using Fully Homomorphic Encryption. PPAI-24@The 38th Annual AAAI Conference on Artificial Intelligence 2024. (Accepted).
Prajwal Panzade, Daniel Takabi and Zhipeng Cai. MedBlindTuner: Towards Privacy-preserving Fine-tuning on Biomedical Images with Transformers and Fully Homomorphic Encryption. W3PHIAI-24@The 38th Annual AAAI Conference on Artificial Intelligence 2024. (Accepted).
Prajwal Panzade, Daniel Takabi and Zhipeng Cai. Privacy-preserving Machine Learning using Functional Encryption: Opportunities and Challenges. IEEE Internet of Things Journal (IF: 10.6) (supported by Microsoft Research and NSF).
Haddadian G, Takabi D, Panzade P, Kim M. A Design Study of Problem-centered Instruction (PCI) for Private Artificial Intelligence (AI) Curriculum Development. Association for Educational Communications and Technology Conference 2023 (supported by NSF).
Prajwal Panzade and Daniel Takabi. FENet: Privacy-preserving Neural Network Training with Functional Encryption. IWSPA @ ACM CODASPY 2023. (supported by Microsoft Research and NSF).
Ghazvinian P, Podschwadt R, Panzade P, Rafiei MH, Takabi D. Poster: Packing-aware Pruning for Efficient Private Inference based on Homomorphic Encryption. IEEE Symposium on Security & Privacy 2023
Prajwal Panzade and Daniel Takabi. Poster: Privacy-preserving Neural Network with Functional Encryption. IEEE Symposium on Security & Privacy 2022
Prajwal Panzade and Daniel Takabi. SoK: Privacy Preserving Machine Learning using Functional Encryption: Opportunities and Challenges.
Prajwal Panzade and Daniel Takabi. Towards Faster Functional Encryption for Privacy-preserving Machine Learning. IEEE International Conference on Trust, Privacy, and Security in Intelligent Systems, and Applications, Dec 2021. IEEE (supported by Microsoft Research and NSF).
Panzade PP, Prakash CS, Maheshkar S, Om H. Detection of Copy-move forgery using AKAZE and SIFT keypoint extraction. Journal on Multimedia Tools and Applications. May 2019. Springer.
Panzade PP, Prakash CS, Maheshkar S. Copy-move forgery detection by using HSV preprocessing and keypoint extraction. In Parallel, Distributed and Grid Computing (PDGC), 2016 Fourth International Conference on 2016 Dec 22 (pp. 264-269). IEEE.
Geethika S, Sai Sree I, Pranathi A, Prajwal P, Advertisements and Multimedia Recommendation based on Age, Gender and Emotion (AMRAGE). ICCCMLA 2019: International Conference on Cybernetics, Cognition, and Machine Learning Applications. Springer.