EMC2 - Energy Efficient Machine Learning and Cognitive Computing

memory Upcoming Workshops

description Workshop Objective

Artificial intelligence (AI) continues to proliferate everyday life aided by the advances in automation, algorithms, and innovative hardware and software technologies. With the growing prominence of AI, Multimodal Large Language model (MLLM) has been rising as new foundational model architecture that uses powerful Large Language Model (LLM) with multimodal tasks efficiency. MLLM are able to achieve surprising capabilities based on text and images, suggesting path for Artificial General Intelligence (AGI). With the advent of new frontiers on execution of MLLM, we are facing new challenges on the ecosystem of software/hardware co-design. There is a realization about the energy cost of developing and deploying MLLM. Training and inferencing the most successful MLLM models has become exceedingly power-hungry often dwarfing the energy needs of entire households for years. At the edge, applications which use these LLMs models for inference are ubiquitous in cell phones, appliances, smart sensors, vehicles, and even wildlife monitors where efficiency is paramount for practical reasons.

format_list_bulleted Topics for the Workshop

  • Neural network architectures for resource constrained applications
  • Efficient hardware designs to implement neural networks including sparsity, locality, and systolic designs
  • Power and performance efficient memory architectures suited for neural networks
  • Network reduction techniques – approximation, quantization, reduced precision, pruning, distillation, and reconfiguration
  • Exploring interplay of precision, performance, power, and energy through benchmarks, workloads, and characterization
  • Simulation and emulation techniques, frameworks, tools, and platforms for machine learning
  • Optimizations to improve performance of training techniques including on-device and large-scale learning
  • Load balancing and efficient task distribution, communication and computation overlapping for optimal performance
  • Verification, validation, determinism, robustness, bias, safety, and privacy challenges in AI systems

Recent Editions

View the complete list of previous editions.