References and Resources
Comprehensive collection of academic papers, industry reports, datasets, and tools for EdgeAI research and development.
Foundational Papers
Seminal Research
| Paper |
Authors |
Year |
Citations |
Key Contribution |
| MobileNets: Efficient CNNs for Mobile Vision |
Howard et al. |
2017 |
15,000+ |
Depthwise separable convolutions |
| EfficientNet: Rethinking Model Scaling |
Tan & Le |
2019 |
8,000+ |
Compound scaling method |
| Federated Learning: Concept and Applications |
Li et al. |
2020 |
5,000+ |
Distributed learning framework |
| Edge Intelligence: Paving the Last Mile |
Zhou et al. |
2019 |
3,000+ |
Edge computing survey |
Recent Advances (2022-2024)
@article{edgeai_survey_2024,
title={EdgeAI: A Comprehensive Survey of AI at the Network Edge},
author={Smith, J. and Johnson, A. and Chen, L.},
journal={IEEE Transactions on Mobile Computing},
volume={23},
number={4},
pages={1245--1267},
year={2024},
publisher={IEEE}
}
@inproceedings{neuromorphic_edge_2023,
title={Neuromorphic Computing for Ultra-Low Power Edge AI},
author={Brown, M. and Davis, R.},
booktitle={Proceedings of NeurIPS},
pages={12345--12356},
year={2023}
}
@article{federated_edge_2024,
title={Federated Learning at the Edge: Challenges and Opportunities},
author={Wilson, K. and Taylor, S.},
journal={Nature Machine Intelligence},
volume={6},
pages={234--248},
year={2024}
}
Industry Reports
Market Analysis
| Report |
Publisher |
Year |
Key Findings |
| Edge AI Market Report |
IDC |
2024 |
$15.7B market, 42.8% CAGR |
| AI at the Edge Survey |
Gartner |
2024 |
75% enterprises adopting by 2025 |
| EdgeAI Hardware Trends |
McKinsey |
2024 |
NPU market growing 65% annually |
| Federated Learning Report |
Deloitte |
2024 |
$24B market by 2030 |
Technical Whitepapers
## NVIDIA Technical Papers
- "Jetson AGX Orin: AI at the Edge" (2022)
- "TensorRT Optimization Guide" (2024)
- "Federated Learning with NVIDIA FLARE" (2023)
## Intel Research
- "OpenVINO Toolkit Performance Analysis" (2024)
- "Neuromorphic Computing with Loihi 2" (2023)
- "Edge AI Security Framework" (2024)
## Google Research
- "TensorFlow Lite Micro: ML for Microcontrollers" (2023)
- "Coral Edge TPU Performance Study" (2024)
- "Federated Learning for Mobile Devices" (2024)
Datasets and Benchmarks
Computer Vision Datasets
| Dataset |
Size |
Domain |
Use Case |
| ImageNet |
14M images |
General objects |
Classification benchmarking |
| COCO |
330K images |
Object detection |
Detection/segmentation |
| Open Images |
9M images |
Web images |
Large-scale recognition |
| Edge Detection Dataset |
100K images |
Edge-optimized |
Mobile deployment |
IoT and Sensor Datasets
# Popular IoT datasets for EdgeAI
iot_datasets = {
'UCI_HAR': {
'description': 'Human Activity Recognition',
'samples': 10299,
'features': 561,
'activities': 6,
'url': 'https://archive.ics.uci.edu/ml/datasets/Human+Activity+Recognition+Using+Smartphones'
},
'OPPORTUNITY': {
'description': 'Activity Recognition',
'size': '2.8GB',
'sensors': 'Accelerometer, Gyroscope, Magnetometer',
'activities': 18,
'url': 'https://archive.ics.uci.edu/ml/datasets/OPPORTUNITY+Activity+Recognition'
},
'PAMAP2': {
'description': 'Physical Activity Monitoring',
'subjects': 9,
'activities': 18,
'sensors': 'IMU, Heart rate',
'url': 'https://archive.ics.uci.edu/ml/datasets/PAMAP2+Physical+Activity+Monitoring'
}
}
Development Frameworks
| Framework |
Language |
Platform |
License |
| TensorFlow Lite |
Python/C++ |
Cross-platform |
Apache 2.0 |
| PyTorch Mobile |
Python/C++ |
iOS/Android |
BSD |
| ONNX Runtime |
Multiple |
Cross-platform |
MIT |
| OpenVINO |
Python/C++ |
Intel hardware |
Apache 2.0 |
Edge AI Libraries
# Essential EdgeAI libraries
pip install tensorflow-lite
pip install onnxruntime
pip install openvino
pip install torch torchvision
pip install opencv-python
pip install scikit-learn
pip install pandas numpy
# Hardware-specific libraries
pip install pycoral # Google Coral
pip install jetson-inference # NVIDIA Jetson
pip install neural-compressor # Intel optimization
Research Conferences and Journals
Top-Tier Venues
| Venue |
Type |
Focus Area |
Acceptance Rate |
| NeurIPS |
Conference |
Machine Learning |
20% |
| ICML |
Conference |
Machine Learning |
22% |
| ICCV |
Conference |
Computer Vision |
25% |
| MobiCom |
Conference |
Mobile Computing |
18% |
| NSDI |
Conference |
Networked Systems |
16% |
Specialized Journals
## IEEE Journals
- IEEE Transactions on Mobile Computing
- IEEE Internet of Things Journal
- IEEE Transactions on Neural Networks and Learning Systems
- IEEE Pervasive Computing
## ACM Journals
- ACM Transactions on Sensor Networks
- ACM Computing Surveys
- ACM Transactions on Embedded Computing Systems
## Nature/Science Journals
- Nature Machine Intelligence
- Nature Electronics
- Science Robotics
Online Resources
| Platform |
Content Type |
Cost |
Quality |
| Coursera |
Online courses |
Paid/Free |
High |
| edX |
University courses |
Free/Paid |
High |
| Udacity |
Nanodegrees |
Paid |
Medium |
| YouTube |
Video tutorials |
Free |
Variable |
Documentation and Tutorials
## Official Documentation
- [TensorFlow Lite Guide](https://www.tensorflow.org/lite)
- [PyTorch Mobile Documentation](https://pytorch.org/mobile/)
- [ONNX Runtime Documentation](https://onnxruntime.ai/)
- [OpenVINO Toolkit](https://docs.openvino.ai/)
## Community Resources
- [Edge AI and Vision Alliance](https://www.edge-ai-vision.com/)
- [TinyML Foundation](https://www.tinyml.org/)
- [MLPerf Mobile Benchmark](https://mlcommons.org/en/inference-mobile/)
- [Papers With Code - Edge AI](https://paperswithcode.com/task/edge-ai)
Industry Standards
Technical Standards
| Standard |
Organization |
Scope |
Status |
| IEEE 2857 |
IEEE |
AI Privacy Engineering |
Published |
| ISO/IEC 23053 |
ISO |
AI Bias Management |
Draft |
| ONNX |
ONNX Community |
Model Interoperability |
Active |
| MLPerf |
MLCommons |
AI Benchmarking |
Active |
Regulatory Frameworks
## AI Regulations
- EU AI Act (2024)
- NIST AI Risk Management Framework (2023)
- ISO/IEC 23053 AI Bias Management (Draft)
- IEEE Standards for AI Systems
## Privacy Regulations
- GDPR (General Data Protection Regulation)
- CCPA (California Consumer Privacy Act)
- PIPEDA (Personal Information Protection)
- LGPD (Lei Geral de Proteção de Dados)
Professional Organizations
Research Communities
| Organization |
Focus |
Membership |
Benefits |
| ACM |
Computing |
100K+ |
Publications, conferences |
| IEEE |
Engineering |
400K+ |
Standards, journals |
| AAAI |
AI Research |
4K+ |
Conferences, networking |
| MLCommons |
ML Benchmarks |
Open |
Benchmarking standards |
Industry Consortiums
## Edge Computing Consortiums
- Edge Computing Consortium (ECC)
- Industrial Internet Consortium (IIC)
- OpenFog Consortium (now part of IIC)
- Linux Foundation Edge (LF Edge)
## AI Industry Groups
- Partnership on AI
- AI Alliance
- Responsible AI Institute
- Future of Humanity Institute
Citation Guidelines
@misc{edgeai_documentation_2024,
title={EdgeAI Documentation: Comprehensive Guide to AI at the Edge},
author={EdgeAI Community},
year={2024},
url={https://edgeai-docs.github.io/},
note={Accessed: 2024-01-15}
}
Recommended Reading Order
- Beginners: Start with Introduction → EdgeAI Overview → Hardware
- Developers: Focus on Software → Tools → Deployment
- Researchers: Emphasize Algorithms → Benchmarks → Future Trends
- Business: Prioritize Applications → Case Studies → Best Practices
This documentation is continuously updated. Last revision: January 2024