公开(公告)号：WO2018208939A1

公开(公告)日：2018-11-15

申请号：PCT/US2018/031833

申请日：2018-05-09

Applicant: NEURALA, INC.

Inventor： LUCIW, Matthew ,  OLIVERA, Santiago ,  GORSHECHNIKOV, Anatoly ,  WURBS, Jeremy ,  VERSACE, Heather Ames ,  VERSACE, Massimiliano

IPC: G06K9/46 ,  G06K9/62 ,  G06K9/00 ,  G06K9/36 ,  G06K9/48

Abstract: Lifelong Deep Neural Network (L-DNN) technology revolutionizes Deep Learning by enabling fast, post-deployment learning without extensive training, heavy computing resources, or massive data storage. It uses a representation-rich, DNN-based subsystem (Module A) with a fast-learning subsystem (Module B) to learn new features quickly without forgetting previously learned features. Compared to a conventional DNN, L-DNN uses much less data to build robust networks, dramatically shorter training time, and learning on-device instead of on servers. It can add new knowledge without re-training or storing data. As a result, an edge device with L-DNN can learn continuously after deployment, eliminating massive costs in data collection and annotation, memory and data storage, and compute power. This fast, local, on-device learning can be used for security, supply chain monitoring, disaster and emergency response, and drone-based inspection of infrastructure and properties, among other applications.

公开(公告)号：WO2014204615A2

公开(公告)日：2014-12-24

申请号：PCT/US2014/039162

申请日：2014-05-22

Applicant: NEURALA, INC.

Inventor： GORSHECHNIKOV, Anatoly ,  VERSACE, Massimiliano ,  VERSACE, Heather Ames ,  LIVITZ, Gennady

IPC: G06F9/315

CPC classification number: H04L67/125 ,  G06F9/5066 ,  G06F9/546 ,  G06N3/0454 ,  H04L43/045 ,  H04L67/10

Abstract: The system and methods disclosed herein include a runtime architecture that takes a nonspecific set of systems of differential equations, distributes them across the network, and iteratively integrates them through time with a possibility to output the results on every iteration. Embodiments of the disclosed system may be used for neural computation or any other suitable application. Embodiments can be used as a standalone engine or as part of another computational system for massively parallel numerical integration of a data-driven dynamical system.

Abstract translation: 本文公开的系统和方法包括运行时架构，其采用非特定的微分方程组系统，将其分布在网络上，并且通过时间迭代地将它们整合，并且可以在每次迭代时输出结果。 所公开的系统的实施例可以用于神经计算或任何其它合适的应用。 实施例可以用作独立的引擎，或者作为用于数据驱动的动态系统的大规模并行数字集成的另一计算系统的一部分。

公开(公告)号：WO2018170512A1

公开(公告)日：2018-09-20

申请号：PCT/US2018/023155

申请日：2018-03-19

Applicant: NEURALA, INC. ,  NEVES, Lucas ,  DEBEASI, Liam ,  VERSACE, Heather Ames

Inventor： NEVES, Lucas ,  DEBEASI, Liam ,  VERSACE, Heather Ames ,  WURBS, Jeremy ,  GORSHECHNIKOV, Anatoly ,  VERSACE, Massimiliano ,  KATZ, Warren

IPC: G06F17/00 ,  G06F17/30 ,  G06K9/00

Abstract: Today, artificial neural networks are trained on large sets of manually tagged images. Generally, for better training, the training data should be as large as possible. Unfortunately, manually tagging images is time consuming and susceptible to error, making it difficult to produce the large sets of tagged data used to train artificial neural networks. To address this problem, the inventors have developed a smart tagging utility that uses a feature extraction unit and a fast-learning classifier to learn tags and tag images automatically, reducing the time to tag large sets of data. The feature extraction unit and fast-learning classifiers can be implemented as artificial neural networks that associate a label with features extracted from an image and tag similar features from the image or other images with the same label. Moreover, the smart tagging system can learn from user adjustment to its proposed tagging. This reduces tagging time and errors.