On-device transfer learning is crucial for adapting a common backbone model to the unique environment of each edge device. Tiny microcontrollers, such as the Raspberry Pi Pico, are key targets for on-device learning but often lack floating-point units, necessitating integer-only training. Dynamic computation of quantization scale factors, which is adopted in former studies, incurs high computational costs. Therefore, this study focuses on integer-only training with static scale factors, which is challenging with existing training methods. We propose a new training method named PRIOT, which optimizes the network by pruning selected edges rather than updating weights, allowing effective training with static scale factors. The pruning pattern is determined by the edge-popup algorithm, which trains a parameter named score assigned to each edge instead of the original parameters and prunes the edges with low scores before inference. Additionally, we introduce a memory-efficient variant, PRIOT-S, which only assigns scores to a small fraction of edges. We implement PRIOT and PRIOT-S on the Raspberry Pi Pico and evaluate their accuracy and computational costs using a tiny CNN model on the rotated MNIST dataset and the VGG11 model on the rotated CIFAR-10 dataset. Our results demonstrate that PRIOT improves accuracy by 8.08 to 33.75 percentage points over existing methods, while PRIOT-S reduces memory footprint with minimal accuracy loss.

The accuracy of tinyML applications is often affected by various environmental factors, such as noises, location/calibration of sensors, and time-related changes. This article introduces a neural network based on-device learning (ODL) approach to address this issue by retraining in deployed environments. Our approach relies on semi-supervised sequential training of multiple neural networks tailored for low-end edge devices. This article introduces its algorithm and implementation on wireless sensor nodes consisting of a Raspberry Pi Pico and low-power wireless module. Experiments using vibration patterns of rotating machines demonstrate that retraining by ODL improves anomaly detection accuracy compared with a prediction-only deep neural network in a noisy environment. The results also show that the ODL approach can save communication cost and energy consumption for battery-powered Internet of Things devices.

Suppose you are alone at home or out for shopping or on vacations and someone breaks into your house. First thing comes into your mind is: if there is some gadget or home security system which can alert you or your neighbors. The home security camera does a good job but they may not work in complete dark. Also, you do not want your gadget to turn on false alarm if it is a cat. In this project I built a proof of concept which merely turn on an LED for demonstration purpose when it detects a person in light or dark using just a Raspberry Pi Pico and a low resolution thermal camera.

The Raspberry Pi Pico wouldn't be the first board that comes to mind for machine learning, but it seems that the $4 may be a viable platform for machine learning projects. The Pico4ML from Arducam is an RP2040 based board with an onboard camera, screen, and microphone that looks to be the same size as the Raspberry Pi Pico. Arducam is probably better known for its range of cameras for the Raspberry Pi and Nvidia Jetson boards, but since the release of the Raspberry Pi Pico, they have been tinkering with machine learning projects powered by the Pico. The Arducam Pico4ML is their first RP2040-based board and the first board to feature an onboard camera, a microphone that you can use for "wake word" detection, a screen, and an Inertial Measurement Unit (IMU) that can detect gestures. The Pico4ML is intended for machine learning and artificial intelligence projects based around Tiny Machine Learning (TinyML).

At the recent tinyML Summit 2021, Raspberry Pi co-founder Eben Upton teased the future of'Pi Silicon' and it looks like machine learning could see a massive improvement thanks to Raspberry Pi's news in-house chip development team. It is safe to say that the Raspberry Pi Pico and its RP2040 SoC have been popular. The Pico has only been on the market for a few weeks, but already has sold 250,000 units with 750,000 on back order. There is a need for more boards powered by the RP2040 and partners such as Adafruit, Pimoroni, Adafruit and Sparkfun are releasing their own hardware, many with features not found on the Pico. Raspberry Pi's in house application specific integrated circuit (ASIC) team are working on the next iteration, and seems to be focused on lightweight accelerators for ultra low power machine learning applications.