부산대학교 도서관

This paper introduces a novel framework that melds the profound tenets of Indian philosophy with AI and ML. Our goal is to create models that replicate human cognition, paving the way for robots that think like us. Key concepts from Indian philosophy, including ”buddhi,” ”manas,” ”chit,” and ”ahamkar,” are woven into the framework, enhancing AI’s ability to understand and interact with the world in a more human-like way. The core of the framework consists of ”Buddhi” and ”Manas.” Buddhi, a discriminative AI component, adeptly navigates sensory data, making decisions aligned with human discernment. Complementing it, ”Manas” is a generative core, capable of crafting narratives and generating creative solutions. This synergy creates an iterative loop that encapsulates human thought, enabling machines to think, visualize, and even dream. In addition to advancing AI and ML research, we conducted an experiment using a modified News Aggregator dataset from the UCI repository. The experiment focused on word prediction, and our model, infused with Indian philosophy, outperformed RNN and LSTM models over 30 epochs. This framework has the potential to revolutionize AI and create more human-like machines, ushering in a new era of AI and robotics.