Overview
The Kaspersky Neuromorphic Platform (KNP) is a software platform for developing, training, and executing Spiking Neural Networks (SNNs). It is designed to work with various types of input data, including telemetry, events, images, 3D data, audio, and tactile data. KNP allows users to create and train SNNs, convert Artificial Neural Networks (ANNs) into SNNs, and optimize network structures.
The platform supports application development in both C++ and Python and can be deployed on standard CPUs as well as the AltAI-1 neuromorphic processor. This makes it suitable for a range of applications, from research into new neural network topologies and learning rules to implementing solutions for robotics, IoT, and other low-power devices.
Key features include:
Creation and training of SNNs.
Conversion of ANNs to SNNs.
Development of new neuron models and synaptic plasticity rules.
Support for both CPU and neuromorphic hardware (AltAI-1).
Aimed at applications in robotics, IoT, unmanned systems, and wearable devices.
