The complex nature of real-world problems calls for heterogeneity in both machine learning (ML) models and hardware systems. The heterogeneity in ML models comes from multi-sensor perceiving and multi-task learning, i.e., multi-modality multi-task (MMMT), resulting in diverse deep neural network (DNN) layers and computation patterns. The heterogeneity in systems comes from diverse processing components, as it becomes the prevailing method to integrate multiple dedicated accelerators into one system. Therefore, a new problem emerges: heterogeneous model to heterogeneous system mapping (H2H). While previous mapping algorithms mostly focus on efficient computations, in this work, we argue that it is indispensable to consider computation and communication simultaneously for better system efficiency. We propose a novel H2H mapping algorithm with both computation and communication awareness; by slightly trading computation for communication, the system overall latency and energy consumption can be largely reduced. The superior performance of our work is evaluated based on MAESTRO modeling, demonstrating 15%-74% latency reduction and 23%-64% energy reduction compared with existing computation-prioritized mapping algorithms.
Heterogeneous Graph Representation Learning with Relation Awareness
Beihang University(北京航空航天大学);the University of Tennessee, Knoxville(田纳西大学诺克斯维尔分校);Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));Hong Kong University of Science and Technology(香港科技大学)
CAT: Can Trust be Predicted with Context-Awareness in Dynamic Heterogeneous Networks?
State Key Laboratory of Integrated Services Networks, School of Cyber Engineering, Xidian University(信息服务网络国家重点实验室,电子科技大学信息工程学院);Hangzhou Institute of Technology, Xidian University(西安电子科技大学杭州研究院);Department of Computer Science, Purdue University(计算机科学系,普渡大学)
AMP: Automatically Finding Model Parallel Strategies with Heterogeneity Awareness
Carnegie Mellon University(卡内基梅隆大学);Mohamed Bin Zayed University of Artificial Intelligence(穆罕默德·本·扎耶德人工智能大学);Petuum Inc.(Petuum公司);University of California, Berkeley(加州大学伯克利分校)
VISTA: An Attention-Based Multi-Agent Reinforcement Learning Architecture for Space Situational Awareness Sensor Tasking
ETSIAE-School of Aeronautics, Universidad Politécnica de Madrid(马德里理工大学航空学院);Indra Sistemas S.A.(英德拉系统公司);Escuela de Ingeniería de Fuenlabrada, Universidad Rey Juan Carlos(胡安卡洛斯国王大学丰拉夫拉达工程学院);Department of Computer Systems Engineering, Universidad Politécnica de Madrid(马德里理工大学计算机系统工程系)
Learning Multi-Task Gaussian Process Over Heterogeneous Input Domains
School of Energy and Power Engineering, Dalian University of Technology(大连理工大学能源与动力工程学院);School of Computer Science and Engineering, Nanyang Technological University(南洋理工大学计算机科学与工程学院)