Paper Presentation: DL-Backtrace by AryaXAI: A Model Agnostic Explainability for Deep Learning Models

​Paper Presentation

​DL-Backtrace by AryaXAI: A Model Agnostic Explainability for Deep Learning Models

​- By Pratinav Seth, Research Scientist, AryaXAI

​Developed by AryaXAI, DL-Backtrace is a pioneering approach designed to redefine how we interpret and optimize deep learning systems. Going beyond typical explanations, DL-Backtrace offers a model-agnostic approach to dissecting the inner workings of neural networks, providing insights into how these models perceive data and make decisions. With DL-Backtrace, you can explain model outcomes in detail and measure each adjustment's impact, allowing for precision in fine-tuning, targeted improvements, and weight redistribution to align outcomes with business objectives.

​In this session, Pratinav Seth, Research Scientist at AryaXAI, will present the core principles of DL-Backtrace and demonstrate its application in enhancing transparency for deep learning models across various use cases.

​We will conclude with a dedicated Q&A session for participants to ask questions, engage with the speaker, and discuss the research findings in depth.

​Meet our Speaker:

​Pratinav Seth

​Pratinav Seth is a research scientist at Arya.ai, where he works on enhancing the reliability and interpretability of AI models. His expertise spans multiple AI domains, with a particular emphasis on creating resource-efficient models and promoting AI trustworthiness through explainability and uncertainty quantification. A recent graduate of the Manipal Institute of Technology, he was recognized as an AAAI Undergraduate Consortium Scholar in 2023. His research has been showcased at prestigious conferences, including CVPR and NeurIPS.

​This session is part of AI Paper-fest 2024 by The SSI Club. For more information and to register for other presentations, visit papers.ssiclub.ai