NeuroMem neuromorphic chip

Trainable, Responsible, Explainable AI
  • Digital neural network technology inspired by biology
  • A learning and inference engine in the network
  • Detailed outputs including novelty and uncertainty flags
  • The knowledge built by the neurons is traceable
  • Deterministic latencies
  • IP available on FPGA and ASIC
  • Simple RTL interface through 15 registers
  • Libraries and apps to get you started
  • Easy to train, validate and deploy

Two powerful classifiers in a same IP core

Radial Basis Function (RBF)

  • Highly adaptive model generator
  • Powerful non linear classifier
  • Notion of positive and uncertain classification, essential for robust decision making
  • Notion of unknown, essential for prediction and learning causality

 

K-Nearest Neighbor (KNN)

  • A classification mode only! Not a model generator
  • Equivalent to “RBF”, only throwing a dice in lieu of reporting an “Unknown”
  • Always retun a response, but the NEAREST can still be far
  • Commonly used for clustering algorithms

Parallel architecture

The NeuroMem homogeneous and parallel architecture allows the design of low-power and high-speed AI chips with unlimited neural network capacity.

  • Neuron cell = memory + processing logic
  • Neuron cells are identical and interconnected
  • Neurons are accessed simultaneously in broadcast mode
  • Neurons work in parallel without the need for a supervisor

A proven technology

1993

  • ZISC
  • IBM
  • 36 and 78 neurons

2007

  • CM1K
  • General Vision
  • 1024 neurons

2012

2015

  • QuarkSE
  • Intel Corporation
  • 128 neurons inside

2017

  • NM500
  • Nepes Corporation
  • 576 neurons

Not just words...

Artificial Intelligence in Healthcare Market Analysis Report - 2020-2026

  • By Size, Share and Growth; By companies, region, type and end-use industry.
  • Key Players – Intel, IBM, Google, Microsoft, General Vision

Self-Learning Neuromorphic Chip Market – Forecast 2023

  • By Application (Image Recognition, Signal Recognition, Data Mining); By Vertical (Healthcare, Power & Energy, Automotive, Media & Entertainment, Aerospace & Defense, Smartphones, Consumer Electronics, Others) 
  • Key players – IBM, Qualcomm , HRL Laboratories , General Vision, Numenta , Hewlett-Packard , Samsung Group , Intel Corporation, Applied Brain Research Inc. , Brainchip Holdings Ltd..

Neuromorphic Computing Market Analysis - 2017-2023

  • By Application (Image Processing, Signal Processing, Data Mining). By End User (Consumer Electronics, Automotive, Military & Defense, Healthcare).
  • Key players - IBM, Hewlett Packard Enterprise Company, Samsung Electronics Co. Ltd., Intel , Qualcomm , General Vision, Brain Corporation, Vicarious, Knowm Inc., and Numenta.

Neuromorphic Computing Market - Global Forecast to 2022

  • By Offering (Hardware, Software); By Application (Image Recognition, Signal Recognition, Data Mining); By Industry (Aerospace & Defense, IT & Telecom, Automotive, Medical & Industrial) and Geography.
  • Key players - IBM, HP Corp., Samsung Electronics Ltd., Intel , HRL Laboratories, General Vision, Applied Brain Research, and BrainChip.