Invention Grant
- Patent Title: Efficient and scalable systems for calculating neural network connectivity in an event-driven way
-
Application No.: US14873138Application Date: 2015-10-01
-
Publication No.: US10339439B2Publication Date: 2019-07-02
- Inventor: Mikko H. Lipasti , Andrew Nere , Atif Hashmi , John F. Wakerly
- Applicant: Thalchemy Corporation
- Applicant Address: US WI Madison
- Assignee: Thalchemy Corporation
- Current Assignee: Thalchemy Corporation
- Current Assignee Address: US WI Madison
- Agency: Lowenstein Sandler LLP
- Main IPC: G06N3/04
- IPC: G06N3/04 ; G06N3/08

Abstract:
Systems and methods achieving scalable and efficient connectivity in neural algorithms by re-calculating network connectivity in an event-driven way are disclosed. The disclosed solution eliminates the storing of a massive amount of data relating to connectivity used in traditional methods. In one embodiment, a deterministic LFSR is used to quickly, efficiently, and cheaply re-calculate these connections on the fly. An alternative embodiment caches some or all of the LFSR seed values in memory to avoid sequencing the LFSR through all states needed to compute targets for a particular active neuron. Additionally, connections may be calculated in a way that generates neural networks with connections that are uniformly or normally (Gaussian) distributed.
Public/Granted literature
- US20160098629A1 EFFICIENT AND SCALABLE SYSTEMS FOR CALCULATING NEURAL NETWORK CONNECTIVITY IN AN EVENT-DRIVEN WAY Public/Granted day:2016-04-07
Information query