Advancements in AI model compression and hardware acceleration are essential for enabling efficient, high-performance deep learning systems. Torch2Chip is a customizable deep neural network compression and deployment toolkit designed to bridge the gap between AI algorithms and prototype hardware accelerators. Developed by Jian Meng, a Ph.D. candidate at Cornell Tech under the guidance of Professor […]
Fusion3D: Research Spotlight
Fusion3D, developed by CoCoSys scholars Sixu Li, Yang (Katie) Zhao, Chaojian Li, Zhifan Ye, and other members of PI Prof. Yingyan (Celine) Lin’s lab at Georgia Tech, presents an end-to-end acceleration framework for real-time 3D intelligence. By accelerating computation and data movement across the algorithm, architecture, and system integration levels, this innovative approach significantly improves […]
Indiana Economic Development Corporation (IEDC) — “Indiana. For the Bold.” — podcast interview with Vijay Raghunatha
Vijay Raghunathan joined Indiana Economic Development Corporation (IEDC)’s Global Economic Summit in May 2024 to discuss Purdue’s pioneering role in semiconductor education and its profound economic impact on Indiana.
Priya Panda receives the inaugural Purdue Engineering 38 by 38 award for remarkable contributions to neuromorphic computing and efficient machine intelligence.
Priya Panda received Purdue’s College of Engineering 38 by 38 award, which recognizes 38 young alumni who have rocketed through the ranks and left their mark by the time they are 38 years old. Priya Panda is recognized for her contributions to the field of machine intelligence have made a significant impact, as evidenced by […]
Akshat Ramachandran and Zishen Wan win second and third place respectively at ACM Student Research Competition
CoCoSys PhD students Akshat Ramachandran and Zishen Wan, both advised by PI Tushar Krishna won second and third position respectively at the ACM Student Research Competition held at MICRO 2024. Their respective papers were “MicroScopiQ: Accelerating Foundational Models through Outlier-Aware Microscaling Quantization” and “Demystifying Neuro-Symbolic AI through Workload Characterization and Software-Hardware Co-Design.”
Purdue University wins Microelectronics Commons Project to advance AI hardware through the Silicon Crossroads Microelectronics Commons Hub
A Purdue University-led team (PI: Kaushik Roy) is the winner of a new project to advance artificial intelligence hardware through the Microelectronic Commons program in collaboration with the Silicon Crossroads Microelectronics Commons (SCMC) Hub.
Seo’s group published paper in IEEE JXCDC
Seo’s group’s recent paper published at IEEE JXCDC (Special Issue on 3D Logic and Memory for Energy Efficient Computing) investigates digital compute-in-memory (DCIM) macros for various 3D IC architectures considering the carbon footprint of 3D architectures. In-house simulator calculates energy and area based on high-level hardware descriptions and neural network workloads, which is integrated with […]
Researchers Develop New System to Optimize Probabilistic Programming
Researchers from the Massachusetts Institute of Technology, the National Institute for Research in Digital Science and Technology, IBM, and Binghamton University have developed Siren, a new programming language that optimizes probabilistic programming through inference plans. The system allows developers to specify how random variables should be processed, leading to improved performance and accuracy in statistical […]
Heck Honored with Test of Time Award for Groundbreaking Research in Deep Learning for Web Search
The ECE professor’s highly cited 2013 paper has become an integral part of modern web search engines.