The Kaspersky Neuromorphic Platform ("KNP" or "platform") is a software platform for developing, training and executing spiking neural networks on a variety of computers.
You can use Kaspersky Neuromorphic Platform to do the following:
Create spiking neural networks (Spiking Neural Network, SNN) and train these on various types of input data, such as telemetry, events, images, 3D data, sound, and tactile data.
A neural network based on artificial neurons that perform temporal summation of input signals represented by spikes and weighted according to the weights of input connections, and generate an output pulse according to a certain threshold-based activation function.
A neural network based on artificial neurons that perform an instantaneous summation of input signals represented as numbers and weighted according to the weights of input connections, and generate an output numerical signal according to a given activation function.
Optimize the structure of the loaded neural networks.
Conduct applied research in the field of input data classification and other application fields of neural networks.
Develop new neural network topologies, for example, impulse analogs of convolutional neural networks that involve convolution in space and time.
Develop new models of synaptic plasticity.
Implement new neuron models.
Implement application solutions based on neuromorphic spiking neural networks in the field of robotic manipulators, Internet of Things, unmanned systems, human-machine interaction, wearable devices, and optimization planning.
Implement application solutions on devices with low power consumption or using neuromorphic processors.
To solve the above tasks, you can use the C++ and Python® languages. The platform supports CPUs as well as the AltAI-1 neuromorphic processing unit designed for power-efficient execution of neural networks on various classes of smart devices.