tinyML Courses

Mastering the Three Pillars of AI Acceleration: Algorithms, Hardware, and Software

Explore IBM Research's holistic approach to AI acceleration, covering low-precision DNN models, scalable hardware techniques, and efficient software methodologies for optimal performance.

• YouTube
• 28 minutes
• On-Demand
• Free Video

Programmable In-Memory Computing Accelerator with 100 SRAM IMC Macros

Explore programmable in-memory computing accelerators for efficient deep neural network processing, featuring 100 SRAM IMC macros and innovative architecture design for improved energy efficiency.

• YouTube
• 30 minutes
• On-Demand
• Free Video

Dissecting a Low Power AI/ML Edge Application - Noise Suppression

Explore noise suppression in AI/ML edge applications, focusing on energy-efficient implementation for battery-powered devices and software frameworks for optimal user experience.

• YouTube
• 27 minutes
• On-Demand
• Free Video

AnalogML - Analog Inferencing for System-Level Power Efficiency

Explore innovative analog machine learning for ultra-low power audio event detection, enabling always-listening edge devices with improved battery life and system-level efficiency.

• YouTube
• 28 minutes
• On-Demand
• Free Video

On-Device Speech Models Optimization and Deployment for Mobile Hardware

Explore on-device speech model optimization and deployment, covering streaming-aware design, quantization techniques, and benchmarks for popular speech processing model topologies on mobile platforms.

• YouTube
• 26 minutes
• On-Demand
• Free Video

Tiny Models with Big Appetites: Cultivating the Perfect Data Diet for Computer Vision

Explore techniques for curating optimal training datasets and designing sampling strategies for tiny computer vision models, emphasizing data quality and model robustness in real-world environments.

• YouTube
• 21 minutes
• On-Demand
• Free Video

TinyML for All: Full-stack Optimization for Diverse Edge AI Platforms

Full-stack optimization for edge AI platforms, addressing challenges in model compression, neural architecture rebalancing, and design primitives to deploy real-world AI on tiny microcontroller units despite limited resources.

• YouTube
• 34 minutes
• On-Demand
• Free Video

TinyM2Net: A Flexible System Algorithm Co-designed Multimodal Learning Framework for Tiny Devices

Explore a flexible multimodal learning framework for tiny devices, featuring depth separable layers and mixed precision quantization, with applications in COVID detection and object recognition.

• YouTube
• 23 minutes
• On-Demand
• Free Video

Power-of-Two Quantization for Low Bitwidth and Hardware Compliant Neural Networks

Explore power-of-two quantization for efficient neural networks, focusing on low bitwidth and hardware compatibility. Learn about key problems, methods, and results in this cutting-edge research.

• YouTube
• 24 minutes
• On-Demand
• Free Video

Improving Energy Efficiency and Robustness of tinyML Computer Vision Using Log-Gradient Input Images

Explore innovative techniques to enhance energy efficiency and robustness in tinyML computer vision, focusing on log-gradient input images and architecture optimization for microcontrollers.

• YouTube
• 25 minutes
• On-Demand
• Free Video

Delta Keyword Transformer: Bringing Transformers to the Edge Through Dynamically Pruned Multi-Head Self-Attention

Explore cutting-edge techniques for optimizing Transformers in edge computing, focusing on the Delta Keyword Transformer and its dynamic pruning approach for efficient multi-head self-attention.

• YouTube
• 21 minutes
• On-Demand
• Free Video

tinyMAN - Lightweight Energy Manager Using Reinforcement Learning for Energy Harvesting Wearable IoT Devices

Explore reinforcement learning for energy management in IoT wearables, focusing on tinyMAN's lightweight approach to achieve energy-neutral operation in harvesting devices.

• YouTube
• 21 minutes
• On-Demand
• Free Video

TinyML Talks UK - Presenting the Upcoming Tiny Cookbook and the Next In-Person Activities

Explore TinyML with upcoming book insights and UK in-person activities. Learn about machine learning on microcontrollers and engage with experts in this exciting field.

• YouTube
• 47 minutes
• On-Demand
• Free Video

IMU Preintegrated Features for Efficient Deep Inertial Odometry

Explore efficient deep inertial odometry using IMU preintegrated features. Learn about measurement errors, problem statement, evaluation, training, and results in human walking applications.

• YouTube
• 22 minutes
• On-Demand
• Free Video

Distributed On-Sensor Compute System for AR/VR Devices - A Semi-Analytical Simulation Framework

Explore a semi-analytical simulation framework for power estimation in distributed on-sensor compute systems for AR/VR devices, focusing on smart image sensors and workload partitioning.

• YouTube
• 22 minutes
• On-Demand
• Free Video