Maximize the performance of your ML platform with Arm software and tools
summary
Covers using Arm’s machine-learning software libraries and tools that are optimized for Arm hardware to improve ML platform performance on endpoint devices. Focuses on concrete Arm technologies and toolchains for deploying and accelerating ML workloads at the edge.
AI-written summary · about
chapters (12)
- 0:00Intro
- 0:17arm AO Dev Summit
- 0:40Machine Learning and Al Basics
- 5:12Arm NN Integration Options with Neural Network Frameworks
- 7:11Relentless Performance Optimization
- 7:48The Flexible Application of Arm NN and Arm Compute Library Three ways applications can use Arm NN and Compute Library
- 9:04The Network Model Tooling Flow: Concept to Deployment
- 9:41Model Architectures
- 10:37Platform Performance Estimation
- 11:04Model Conditioning
- 12:20Profiling and Debug
- 13:29Looking to the Future - Code Generation (CodeGen) There are two main approaches to ML backend runtime implementations
related talks
42mAI and Machine Learning BoF
28mML on the Edge for Industry 4.0 with Arm Ethos N-78 Neural Processing Unit
06mML for Systems and Systems for ML
37mOptimizing AI and ML with Platform Engineering
24mTools & Techniques for a cost optimized Analytics & AI pipeline
21mGetting the best of on-device ML with ML Kit
10mMachine Learning for mobile
14mArm: One Platform for Agentic AI, from Cloud to the Edge