Edge AI Speaker
IEEE Spectrum: New Neural Network Slashes Sensor-Data Overload
R&D World: Sensor data reimagined: When 90% less data can fuel 100x gains in efficiency in AI projects
Going from raw sensor data to a decision advantage
Lightweight AI for edge computing in space
Lightscline's manifesto
Leveraging channels and augmentation to improve sensor datasets
Working with reduced datasets by exploiting sparsity
Leveraging inherent structure to work with 90% less data
Redundancy in sensor data
Training, re-training, and iterating close to the data source
Low-dimensional structures in big data suffice for several tasks
A 90% smaller footprint enables AI deployment at scale
Redundancy enables upfront sampling instead of dimensionality reduction
Preserving physics of vibration data using post-Nyquist data-driven techniques
How to inform models about the inherent structure in data?
Pruning Neural Network - Concepts
