The approximate computing paradigm advocates for relaxing accuracy goals in applications
to improve energy-efficiency and performance. Recently, this paradigm has been explored …

Recent advances in machine learning (ML) have spotlighted the pressing need for
computing architectures that bridge the gap between memory bandwidth and processing …

摘要片上光互连高带宽, 低延时和低功耗的特点可以有效缓解多核处理器中电互连方式的通信
瓶颈. 片上光路由器是片上光网络设计中必不可少的组成部分. 但随着处理器规模的逐渐增大 …

Traditional DRAM-based main memory systems face several challenges with memory
refresh overhead, high latency, and low throughput as the industry moves towards smaller …

Optical Network-on-Chip (ONoC) has become promising due to its advantages in
transmission bandwidth and delay, compared to the traditional electrical interconnects on …

Addressing challenges regarding Hybrid Optoelectronic Network-on-Chip systems, such as
congestion control, their limited adaptability, and their inability to facilitate optoelectronic co …

摘要针对片上光电混合互连网络(hybrid optoelectronic network-on-chip, HONoC)
拥塞控制与自适应能力差, 无法实现光电联合仿真等问题, 提出一种适用于可重构阵列处理器的 …

Abstract Machine learning applications have become increasingly prevalent over the past
decade across many real-world use cases, from smart consumer electronics to automotive …

The approximate computing paradigm advocates for relaxing accuracy goals in applications
to improve energy-efficiency and performance. Recently, this paradigm has been explored …

Abstract Previously, Photonic Network-on-Chip (PNoC) was proposed to solve the serious
bottlenecks experienced by the traditional Electrical Network-on-Chips (ENoCs) such as …