相关实验视频
Updated: May 31, 2025
11:23
Lensless Fluorescent Microscopy on a Chip
Published on: August 17, 2011
17.6K
一个低功率的通用矩阵乘法加速器,具有稀疏的重量和输出静止数据流
1Research Center for Novel Computing Sensing and Intelligent Processing, Zhejiang Lab, Hangzhou 311100, China.
Micromachines
|January 25, 2025
概括
本研究介绍了一种新的稀疏通用矩阵乘法 (GEMM) 加速器,用于在资源有限的设备上高效的机器学习. 该方法通过优化数据移动和缓冲器利用来提高计算和能源效率.
科学领域:
- 计算机科学 计算机科学
- 机器学习硬件 机器学习硬件
背景情况:
- 一般矩阵乘法 (GEMM) 是计算密集型的,限制其在资源有限的环境中使用.
- 现有的GEMM数据再利用方法无法充分利用数据流动的潜在减少.
研究的目的:
- 开发一个稀疏的GEMM加速器,提高边缘设备上机器学习的效率.
- 为了应对硬件中存储和处理稀疏矩阵的挑战.
主要方法:
- 引入了一个重量和输出静止 (WOS) 数据流和分布式缓冲区架构,用于稀疏的GEMM.
- 为压缩的GEMM开发了一个可适应的映射方案,并为权重开发了一个离线的稀疏度意识的混合策略.
- 实施了一种低成本的稀疏计算方法,使用全球共享的输入来实现高吞吐量.
主要成果:
- 稀疏的GEMM加速器以压缩格式处理矩阵,消除了芯片上的重量和部分总和转移.
- 精益求精的混合策略平衡了缓冲器利用率,并最大限度地减少了不规则重量矩阵的浪费.
- 与现有方法相比,FPGA实验显示了1.73倍更好的计算效率和1.36倍更高的能源效率.
结论:
- 建议的稀疏GEMM加速器有效地减少了数据的移动,并提高了机器学习的硬件效率.
- 新的数据流,缓冲架构和重量管理策略使稀疏矩阵的有效处理成为可能.
- 这项工作为在资源有限的平台上部署计算要求高的机器学习模型提供了可行的解决方案.
相关概念视频
01:20Parallel Processing
144
The brain processes sensory information rapidly due to parallel processing, which involves sending data across multiple neural pathways at the same time. This method allows the brain to manage various sensory qualities, such as shapes, colors, movements, and locations, all concurrently. For instance, when observing a forest landscape, the brain simultaneously processes the movement of leaves, the shapes of trees, the depth between them, and the various shades of green. This enables a quick and...
144
01:30Acceleration Vectors
7.9K
In everyday conversation, accelerating means speeding up. Acceleration is a vector in the same direction as the change in velocity, Δv, therefore the greater the acceleration, the greater the change in velocity over a given time. Since velocity is a vector, it can change in magnitude, direction, or both. Thus acceleration is a change in speed or direction, or both. For example, if a runner traveling at 10 km/h due east slows to a stop, reverses direction, and continues their run at 10 km/h...
7.9K
01:17Ampere-Maxwell's Law: Problem-Solving
529
A parallel-plate capacitor with capacitance C, whose plates have area A and separation distance d, is connected to a resistor R and a battery of voltage V. The current starts to flow at t = 0. What is the displacement current between the capacitor plates at time t? From the properties of the capacitor, what is the corresponding real current?
To solve the problem, we can use the equations from the analysis of an RC circuit and Maxwell's version of Ampère's law.
For the first part of...
To solve the problem, we can use the equations from the analysis of an RC circuit and Maxwell's version of Ampère's law.
For the first part of...
529
01:20Vector Operations
1.1K
Vectors are physical quantities that have both magnitude and direction. The vector operations include addition, subtraction, and scalar multiplication.
A vector multiplied by a scalar value is called scalar multiplication. The result obtained is a new vector with a different magnitude. If the scalar is positive, the direction of the vector remains the same, but if it is negative, the direction of the vector is reversed. For example, the product of the mass and velocity yields the momentum.
A vector multiplied by a scalar value is called scalar multiplication. The result obtained is a new vector with a different magnitude. If the scalar is positive, the direction of the vector remains the same, but if it is negative, the direction of the vector is reversed. For example, the product of the mass and velocity yields the momentum.
1.1K
01:19Distributed Loads
508
Distributed loads are a common type of load that engineers and scientists encounter in various practical situations. Distributed loads often refer to a type of load spread over a surface or a structure and can be modeled as continuous force per unit area.
For example, consider a bookshelf filled with books stacked vertically adjacent to each other. The weight of the books is evenly distributed over the length of the shelf. As a result, the pressure at different locations on the surface of the...
For example, consider a bookshelf filled with books stacked vertically adjacent to each other. The weight of the books is evenly distributed over the length of the shelf. As a result, the pressure at different locations on the surface of the...
508
01:24Fast Decoupled and DC Powerflow
164
The fast decoupled power flow method addresses contingencies in power system operations, such as generator outages or transmission line failures. This method provides quick power flow solutions, essential for real-time system adjustments. Fast decoupled power flow algorithms simplify the Jacobian matrix by neglecting certain elements, leading to two sets of decoupled equations:
164