This is a list of interesting papers and projects about TinyML.
TinyML is awesome.
This is a list of interesting papers, projects, articles and talks about TinyML.
[pdf]
[pdf]
[pdf]
[pdf]
[pdf]
[pdf]
[official code]
[AMC] AutoML for Model Compression and Acceleration on Mobile Devices | [pdf]
[official code]
Mobile Machine Learning Hardware at ARM: A Systems-on-Chip (SoC) Perspective | [pdf]
[HAQ] Hardware-Aware Automated Quantization with Mixed Precision | [pdf]
Efficient and Robust Machine Learning for Real-World Systems | [pdf]
[GesturePod] Gesture-based Interaction Cane for People with Visual Impairments | [pdf]
[YOLO-LITE] A Real-Time Object Detection Algorithm Optimized for Non-GPU Computers | [pdf]
[CMSIS-NN] Efficient Neural Network Kernels for Arm Cortex-M CPUs | [pdf]
Quantizing deep convolutional networks for efficient inference: A whitepaper | [pdf]
[Hello Edge] Keyword Spotting on Microcontrollers | [pdf]
▲ Top |
---|
FastGRNN: A Fast, Accurate, Stable and Tiny Kilobyte Sized Gated Recurrent Neural Network | [pdf]
Image Classification on IoT Edge Devices: Profiling and Modeling| [pdf]
[PROXYLESSNAS] DIRECT NEURAL ARCHITECTURE SEARCH ON TARGET TASK AND HARDWARE |[pdf]
[official code]
Energy Efficient Hardware for On-Device CNN Inference via Transfer Learning | [pdf]
Visual Wake Words Dataset | [pdf]
Compiling KB-Sized Machine Learning Models to Tiny IoT Devices | [pdf]
Reconfigurable Multitask Audio Dynamics Processing Scheme | [pdf]
Pushing the limits of RNN Compression | [pdf]
A low-power end-to-end hybrid neuromorphic framework for surveillance applications | [pdf]
Deep Learning at the Edge | [pdf]
Memory-Driven Mixed Low Precision Quantization For Enabling Deep Network Inference On Microcontrollers | [pdf]
[official code]
[SpArSe] Sparse Architecture Search for CNNs on Resource-Constrained Microcontrollers |[pdf]
[MobileNetV2] Inverted Residuals and Linear Bottlenecks |[pdf]
Latent Weights Do Not Exist: Rethinking Binarized Neural Network Optimization |[pdf]
Low-Power Computer Vision: Status, Challenges, Opportunities |[pdf]
▲ Top |
---|
COMPRESSING RNNS FOR IOT DEVICES BY 15-38X USING KRONECKER PRODUCTS |[pdf]
BENCHMARKING TINYML SYSTEMS: CHALLENGES AND DIRECTION |[pdf]
Lite Transformer with Long-Short Range Attention |[pdf]
[FANN-on-MCU] An Open-Source Toolkit for Energy-Efficient Neural Network Inference at the Edge of the Internet of Things |[pdf]
[TENSORFLOW LITE MICRO] EMBEDDED MACHINE LEARNING ON TINYML SYSTEMS |[pdf]
[AttendNets] Tiny Deep Image Recognition Neural Networks for the Edge via Visual Attention Condensers |[pdf]
[TinySpeech] Attention Condensers for Deep Speech Recognition Neural Networks on Edge Devices |[pdf]
Robust navigation with tinyML for autonomous mini-vehicles |[pdf]
[official code]
[MICRONETS] NEURAL NETWORK ARCHITECTURES FOR DEPLOYING TINYML APPLICATIONS ON COMMODITY MICROCONTROLLERS |[pdf]
[TinyLSTMs] Efficient Neural Speech Enhancement for Hearing Aids |[pdf]
[MCUNet] Tiny Deep Learning on IoT Devices |[pdf]
[official code]
Efficient Residue Number System Based Winograd Convolution | [pdf]
On Front-end Gain Invariant Modeling for Wake Word Spotting | [pdf]
TOWARDS DATA-EFFICIENT MODELING FOR WAKE WORD SPOTTING | [pdf]
Accurate Detection of Wake Word Start and End Using a CNN | [pdf]
[PoPS] Policy Pruning and Shrinking for Deep Reinforcement Learning | [pdf]
Howl: A Deployed, Open-Source Wake Word Detection System | [pdf]
[official code]
[LeakyPick] IoT Audio Spy Detector | [pdf]
On-Device Machine Learning: An Algorithms and Learning Theory Perspective | [pdf]
Leveraging Automated Mixed-Low-Precision Quantization for tiny edge microcontrollers | [pdf]
OPTIMIZE WHAT MATTERS: TRAINING DNN-HMM KEYWORD SPOTTING MODEL USING END METRIC | [pdf]
[RNNPool] Efficient Non-linear Pooling for RAM Constrained Inference | [blog]
[pdf]
[official code]
[Shiftry] RNN Inference in 2KB of RAM |[pdf]
[Once for All] Train One Network and Specialize it for Efficient Deployment |[pdf]
[official code]
A Tiny CNN Architecture for Medical Face Mask Detection for Resource-Constrained Endpoints |[pdf]
Rethinking Generalization in American Sign Language Prediction for Edge Devices with Extremely Low Memory Footprint |[pdf]
[presentation]
[ShadowNet] A Secure and Efficient System for On-device Model Inference |[pdf]
Hardware Aware Training for Efficient Keyword Spotting on General Purpose and Specialized Hardware |[pdf]
Automated facial recognition for wildlife that lack unique markings: A deep learning approach for brown bears |[pdf]
[HyNNA]: Improved Performance for Neuromorphic Vision Sensor based Surveillance
using Hybrid Neural Network Architecture |[pdf]
The Hardware Lottery |[pdf]
MLPerf Inference Benchmark |[pdf]
MLPerf Mobile Inference Benchmark : Why Mobile AI Benchmarking Is Hard and What to Do About It |[pdf]
[TinyRL] Learning to Seek: Tiny Robot Learning for Source Seeking on a Nano Quadcopter |[pdf]
[presentation]
Pushing the Limits of Narrow Precision Inferencing at Cloud Scale with Microsoft Floating Point |[pdf]
[TinyBERT] Distilling BERT for Natural Language Understanding |[pdf]
[Larq] An Open-Source Library for Training Binarized Neural Networks |[pdf]
[presentation]
[official code]
[FedML] A Research Library and Benchmark for Federated Machine Learning |[pdf]
Survey of Machine Learning Accelerators |[pdf]
▲ Top |
---|
[I-BERT] Integer-only BERT Quantization |[pdf]
[TinyTL] Reduce Memory, Not Parameters for Efficient On-Device Learning |[pdf]
[official code]
ON THE QUANTIZATION OF RECURRENT NEURAL NETWORKS |[pdf]
[TINY TRANSDUCER] A HIGHLY-EFFICIENT SPEECH RECOGNITION MODEL ON EDGE DEVICES |[pdf]
LARQ COMPUTE ENGINE: DESIGN, BENCHMARK, AND DEPLOY STATE-OF-THE-ART BINARIZED NEURAL NETWORKS |[pdf]
[LEAF] A LEARNABLE FRONTEND FOR AUDIO CLASSIFICATION |[pdf]
Enabling Large NNs on Tiny MCUs with Swapping |[pdf]
Fixed-point Quantization of Convolutional Neural Networks for Quantized Inference on Embedded Platforms |[pdf]
Estimating indoor occupancy through low-cost BLE devices |[pdf]
[Tiny Eats] Eating Detection on a Microcontroller |[pdf]
[DEVICETTS] A SMALL-FOOTPRINT, FAST, STABLE NETWORK FOR ON-DEVICE TEXT-TO-SPEECH |[pdf]
A 0.57-GOPS/DSP Object Detection PIM Accelerator on FPGA |[pdf]
Rethinking Co-design of Neural Architectures and Hardware Accelerators |[pdf]
Sparsity in Deep Learning: Pruning and growth for efficient inference and training in neural networks |[pdf]
[Apollo] Transferable Architecture Exploration |[pdf]
DEEP NEURAL NETWORK BASED COUGH DETECTION USING BED-MOUNTED ACCELEROMETER MEASUREMENTS |[pdf]
TapNet: The Design, Training, Implementation, and Applications of a Multi-Task Learning CNN for Off-Screen Mobile Input|[pdf]
MEMORY-EFFICIENT SPEECH RECOGNITION ON SMART DEVICES |[pdf]
SWIS - Shared Weight bIt Sparsity for Efficient Neural Network Acceleration |[pdf]
Hardware Aware Training for Efficient Keyword Spotting on General Purpose and Specialized Hardware |[pdf]
Hypervector Design for Efficient Hyperdimensional Computing on Edge Devices |[pdf]
When Being Soft Makes You Tough:A Collision Resilient Quadcopter Inspired by Arthropod Exoskeletons |[pdf]
[TinyOL] TinyML with Online-Learning on Microcontrollers |[pdf]
Quantization-Guided Training for Compact TinyML Models |[pdf]
hls4ml: An Open-Source Codesign Workflow to Empower Scientific Low-Power Machine Learning Devices |[pdf]
Memory-Efficient, Limb Position-Aware Hand Gesture Recognition using Hyperdimensional Computing |[pdf]
Dynamically Throttleable Neural Networks(TNN) |[pdf]
A Comprehensive Survey on Hardware-Aware Neural Architecture Search |[pdf]
An Intelligent Bed Sensor System for Non-Contact Respiratory Rate Monitoring |[pdf]
Measuring what Really Matters: Optimizing Neural Networks for TinyML |[pdf]
Few-Shot Keyword Spotting in Any Language |[pdf]
DOPING: A TECHNIQUE FOR EXTREME COMPRESSION OF LSTM MODELS USING SPARSE STRUCTURED ADDITIVE MATRICES |[pdf]
[OutlierNets] Highly Compact Deep Autoencoder Network Architectures for On-Device Acoustic Anomaly Detection |[pdf]
[TENT] Efficient Quantization of Neural Networks on the tiny Edge with Tapered FixEd PoiNT |[pdf]
A 1D-CNN Based Deep Learning Technique for Sleep Apnea Detection in IoT Sensors |[pdf]
ADAPTIVE TEST-TIME AUGMENTATION FOR LOW-POWER CPU |[pdf]
Compiler Toolchains for Deep Learning Workloads on Embedded Platforms |[pdf]
[ProxiMic] Convenient Voice Activation via Close-to-Mic Speech Detected by a Single Microphone |[pdf]
[Fusion-DHL] WiFi, IMU, and Floorplan Fusion for Dense History of Locations in Indoor Environments |[pdf]
[µNAS] Constrained Neural Architecture Search for Microcontrollers |[pdf]
RaspberryPI for mosquito neutralization by power laser |[pdf]
Widening Access to Applied Machine Learning with TinyML |[pdf]
Using Machine Learning in Embedded Systems |[pdf]
[FRILL] A Non-Semantic Speech Embedding for Mobile Devices |[pdf]
Few-Shot Keyword Spotting in Any Language |[pdf]
MLPerf Tiny Benchmark |[pdf]
Efficient Deep Learning: A Survey on Making Deep Learning Models Smaller, Faster, and Better |[pdf]
AttendSeg: A Tiny Attention Condenser Neural Network for Semantic Segmentation on the Edge |[pdf]
RANDOMNESS IN NEURAL NETWORK TRAINING:CHARACTERIZING THE IMPACT OF TOOLING |[pdf]
TinyML: Analysis of Xtensa LX6 microprocessor for Neural Network Applications by ESP32 SoC |[pdf]
[Keyword Transformer]: A Self-Attention Model for Keyword Spotting |[pdf]
LB-CNN: An Open Source Framework for Fast Training of Light Binary Convolutional Neural Networks using Chainer and Cupy |[pdf]
[Only Train Once]: A One-Shot Neural Network Training And Pruning Framework |[pdf]
[BEANNA]: A Binary-Enabled Architecture for Neural Network Acceleration|[pdf]
A TinyML Platform for On-Device Continual Learning with Quantized Latent Replays |[pdf]
CLASSIFICATION OF ANOMALOUS GAIT USING MACHINE LEARNING TECHNIQUES AND EMBEDDED SENSORS |[pdf]
[MOBILEVIT]: LIGHT-WEIGHT, GENERAL-PURPOSE, AND MOBILE-FRIENDLY VISION TRANSFORMER |[pdf]
[MCUNetV2]: Memory-Efficient Patch-based Inference for Tiny Deep Learning |[pdf]
[LCS]: LEARNING COMPRESSIBLE SUBSPACES FOR ADAPTIVE NETWORK COMPRESSION AT INFERENCE TIME |[pdf]
Feature Augmented Hybrid CNN for Stress Recognition Using Wrist-based Photoplethysmography Sensor |[pdf]
[ANALOGNETS]: ML-HW CO-DESIGN OF NOISE-ROBUST TINYML MODELS AND ALWAYS-ON ANALOG COMPUTE-IN-MEMORY ACCELERATOR |[pdf]
[BSC]: Block-based Stochastic Computing to Enable Accurate and Efficient TinyML |[pdf]
[TiWS-iForest]: Isolation Forest in Weakly Supervised and Tiny ML scenarios |[pdf]
[RadarNet]: Efficient Gesture Recognition Technique Utilizing a Miniature Radar Sensor|[pdf]
The Synergy of Complex Event Processing and Tiny Machine Learning in Industrial IoT |[pdf]
▲ Top |
---|
A Heterogeneous In-Memory Computing Cluster For Flexible End-to-End Inference of Real-World Deep Neural Networks |[pdf]
CFU Playground: Full-Stack Open-Source Framework for Tiny Machine Learning (tinyML) Acceleration on FPGAs |[pdf]
BottleFit: Learning Compressed Representations in Deep Neural Networks for Effective and Efficient Split Computing |[pdf]
[UDC]: Unified DNAS for Compressible TinyML Models |[pdf]
A VM/Containerized Approach for Scaling TinyML Applications |[pdf]
A Fast Network Exploration Strategy to Profile Low Energy Consumption for Keyword Spotting |[pdf]
PocketNN: Integer-only Training and Inference of Neural Networks via Direct Feedback Alignment and Pocket Activations in Pure C++ |[pdf]
[TinyMLOps]: Operational Challenges for Widespread Edge AI Adoption |[pdf]
[Auritus]: An Open-Source Optimization Toolkit for Training and Development of Human Movement Models and Filters Using Earables |[pdf]
|[code]
Enabling Hyperparameter Tuning of Machine Learning Classifiers in Production |[pdf]
TinyOdom: Hardware-Aware Efficient Neural Inertial Navigation |[pdf]
|[code]
Searching for Efficient Neural Architectures for On-Device ML on Edge TPUs |[pdf]
Green Accelerated Hoeffding Tree |[pdf]
tinyRadar: mmWave Radar based Human Activity Classification for Edge Computing |[pdf]
MACHINE LEARNING SENSORS |[pdf]
Evaluating Short-Term Forecasting of Multiple Time Series in IoT Environments |[pdf]
How to train accurate BNNs for embedded systems? |[pdf]
Vildehaye: A Family of Versatile, Widely-Applicable, and Field-Proven Lightweight Wildlife Tracking and Sensing Tags |[pdf]
On-Device Training Under 256KB Memory |[pdf]
DEPTH PRUNING WITH AUXILIARY NETWORKS FOR TINYML |[pdf]
[EdgeNeXt]: Efficiently Amalgamated CNN-Transformer Architecture for Mobile Vision Applications |[pdf]
Tiny Robot Learning: Challenges and Directions for Machine Learning in Resource-Constrained Robots |[pdf]
[POET]: Training Neural Networks on Tiny Devices with Integrated Rematerialization and PagingPOET: Training Neural Networks on Tiny Devices |[pdf]
Two-stage Human Activity Recognition on
Microcontrollers with Decision Trees and CNNs |[pdf]
How to Manage Tiny Machine Learning at Scale – An Industrial Perspective |[pdf
[SeLoC-ML]: Semantic Low-Code Engineering for Machine Learning Applications in Industrial IoT|[pdf]
[IMU2Doppler]: Cross-Modal Domain Adaptation for Doppler-based Activity Recognition Using IMU Data" |[pdf]
[Tiny-HR]: Towards an interpretable machine learning
pipeline for heart rate estimation on edge devices |[pdf]
[Enabling Fast Deep Learning on Tiny Energy-Harvesting IoT Devices]|[pdf]
Extremely Simple Activation Shaping for
Out-of-Distribution Detection |[pdf]
A processing‑in‑pixel‑in‑memory paradigm for resource‑constrained TinyML applications |[pdf]
[tinySNN]: Towards Memory- and Energy-Efficient Spiking Neural Networks |[pdf]
[DeepPicarMicro]: Applying TinyML to Autonomous Cyber Physical Systems |[pdf]
Incremental Online Learning Algorithms Comparison for Gesture and Visual Smart Sensors |[pdf
-[Protean]: An Energy-Efficient and Heterogeneous Platform for
Adaptive and Hardware-Accelerated Battery-free Computing |[pdf
IN-SENSOR & NEUROMORPHIC COMPUTING ARE ALL YOU NEED FOR ENERGY
EFFICIENT COMPUTER VISION |[pdf]
Energy Efficient Hardware Acceleration of
Neural Networks with Power-of-Two
Quantisation |[pdf]
Enabling ISP-less Low-Power Computer Vision |[pdf]
Rethinking Vision Transformers for MobileNet Size and Speed |[pdf]
Neuromorphic Computing and Sensing in Space |[pdf]
Joint Data Deepening-and-Prefetching for Energy-Efficient Edge Learning |[pdf]
PreMa: Predictive Maintenance of Solenoid Valve in Real-Time at Embedded Edge-Level | pdf]
▲ Top |
---|
Exploring Automatic Gym Workouts Recognition Locally On Wearable Resource-Constrained Devices |[pdf]
[MetaLDC]: Meta Learning of Low-Dimensional Computing Classifiers for Fast On-Device Adaption |[pdf]
Faster Attention Is What You Need: A Fast
Self-Attention Neural Network Backbone
Architecture for the Edge via Double-Condensing
Attention Condensers |[pdf]
[TinyReptile]: TinyML with Federated Meta-Learning |[pdf
[TinyProp] - Adaptive Sparse Backpropagation for Efficient TinyML On-device Learning |[pdf
[LiteTrack] - Layer Pruning with Asynchronous Feature Extraction
for Lightweight and Efficient Visual Tracking - Adaptive Sparse Backpropagation for Efficient TinyML On-device Learning |[pdf
▲ Top |
---|
[official code]
[presentation]
[presentation]
2020-09
Autonomous embedded driving using computer vision
2020-10
EleTect - TinyML and IoT Based Smart Wildlife Tracker
2020-03
Handwriting Recognition
2021-01
Why Benchmarking TinyML Systems Is Challenging
2021-01
Build your own Google Assistant using tinyML
2021-02
Fall detection and heart rate monitoring using AVR-IoT
2021-02
The Maker Show: TinyML for wildlife conservation
2021-05
Under $100 and Less Than 1mW: Pneumonia Detection Solution for Everyone
2021-06
Early Pigs' Respiratory Disease Detection Using Edge Impulse
2021-06
Posture Watchdog
2021-07
Localized Environmental Sensing With TinyML
2021
Wireless Quarter: Edge Intelligence
▲ Top |
---|
[GitHub]
[GitHub]
[GitHub]
▲ Top |
---|
[2022-12]
AI at the Edge (D. Situnayake & J. Plunkett, 2022. O'Reilly): [Book]
[2022-10]
Machine Learning on Commodity Tiny Devices (S. Guo & Q. Zhou, 2022. CRC Press): [Book]
[2022-07]
Introduction to TinyML (Rohit Sharma, 2022, AITS): [Book]
| [GitHub]
[2022-04]
TinyML Cookbook (Gian Marco Iodice, 2022. Packt): [Book]
| [GitHub]
[2021-03]
Artificial Intelligence for IoT Cookbook (Michael Roshak, 2021. Packt): [Book]
| [GitHub]
[2020-04]
Mobile Deep Learning with TensorFlow Lite, ML Kit and Flutter: Build scalable real-world projects to implement end-to-end neural networks on Android and iOS (Anubhav Singh, Rimjhim Bhadani, 2020. Packt): [Book]
[2020-01]
TinyML: Machine Learning with TensorFlow Lite on Arduino and Ultra-Low-Power Microcontrollers (Pete Warden. O'Reilly Media): [Book]
▲ Top |
---|
2019-12
TinyML as-a-Service: What is it and what does it mean for the IoT Edge?
2019-12
TinyML as a Service and the challenges of machine learning at the edge
2020-05
Model Quantization Using TensorFlow Lite
2020-09
TinyML is breathing life into billions of devices
2020-12
Predictions for Embedded Machine Learning for IoT in 2021
2020-12
Matthew Mattina: Life-Saving Models in Your Pocket
2020-12
Tiny four-bit computers are now all you need to train AI
2021-01
How predictive maintenance is changing the industrial enterprise for good
2021-02
What is TinyML?
2021-02
How AI is Taking on Sensors
2021-04
MLCommons™ Releases MLPerf™ Inference v1.0 Results with First Power Measurements
2021-05
TapLock - A bike lock with machine learning
2021-05
Taking Back Control
2021-06
Neural network architectures for deploying TinyML applications on commodity microcontrollers
2021-06
TinyML in MicroCosmos
2021-06
‘Small Data’ Are Also Crucial for Machine Learning
2021-07
A natively flexible 32-bit Arm microprocessor
2021-07
Wearable Devices Can Reduce Collision Risk in Blind and Visually Impaired People
2021-09
AI Inspection Using Analog Gauge as an Example
▲ Top |
---|
▲ Top |
---|
[website]
[website]
[website]
[website]
[edx]
[YouTube]
| [edx]
| [GitHub]
[edX]
[Coursera]
[YouTube]
[YouTube]
▲ Top |
---|
[Website]
| [YouTube]
[Website]
| [YouTube]
[Website]
| [YouTube]
Title | Speaker | Published Date | Link |
---|---|---|---|
Challenges for Large Scale Deployment of Tiny ML Devices | G. Raghavan | 2022-04-29 | slide |
Building data-centric AI tooling for embedded engineers | D. Situnayake | 2022-04-29 | slide |
Sensors and ML: waking smarter for less | A. Ataya | 2022-05-04 | slide |
MLOps for TinyML: Challenges & Directions in Operationalizing TinyML at Scale | V.J. Reddi | 2022-05-24 | slide |
Vibration Monitoring Machine Learning Demonstration | J. Edwards | 2020-12-22 | github |
Moving From AI To IntelligentAI To Reduce The Cost Of AI At The Edge | J. Edwards | 2020-12-22 | web |
▲ Top |
---|
[website]
If you have any suggestions about TinyML papers and projects, feel free to mail me :)