RAMAN: Reinforcement Learning Inspired Algorithm for Mapping Applications onto Mesh Network-on-Chip

Article Computer Science, Hardware & Architecture

Core Placement Optimization for Multi-chip Many-core Neural Network Systems with Reinforcement Learning

Nan Wu et al.

Summary: The study proposes a reinforcement-learning-based method to optimize core placement in multi-chip many-core systems, improving system performance and efficiency. Experimental results demonstrate significant improvements in throughput and latency compared to traditional methods.

ACM TRANSACTIONS ON DESIGN AUTOMATION OF ELECTRONIC SYSTEMS (2021)

Add to Collection

Article Computer Science, Hardware & Architecture

An IP Core Mapping Algorithm Based on Neural Networks

Qingkun Chen et al.

Summary: The IP core mapping optimization problem is NP-hard, and the MPNN-Ptr networks and NNMM neural network IP core mapping model are proposed. The NMA algorithm conducts global search and 2-opt algorithm conducts local search, effectively reducing the candidate mapping solution space and improving the mapping quality.

IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS (2021)

Add to Collection

Article Computer Science, Hardware & Architecture

A Reinforcement Learning-Based Framework for Solving the IP Mapping Problem

Qingkun Chen et al.

Summary: The article introduces a reinforcement learning-based framework to enhance heuristic searches through neural network probability models. Experimental results demonstrate that this framework effectively reduces communication costs in IP mapping problems.

IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS (2021)

Add to Collection

Article Computer Science, Information Systems