Massimiliano Versace

US Patent:

20080117220, May 22, 2008

Filed:

Sep 24, 2007

Appl. No.:

11/860254

Inventors:

Anatoli Gorchetchnikov - Belmont MA, US
Heather Marie Ames - South Boston MA, US
Massimiliano Versace - South Boston MA, US
Fabrizio Santini - Jamaica Plain MA, US

Assignee:

Neurala LLC - Boston MA

International Classification:

G06T 1/20
G06F 15/76

US Classification:

345503, 345522, 345536

Abstract:

An accelerator system is implemented on an expansion card comprising a printed circuit board having (a) one or more graphics processing units (GPU), (b) two or more associated memory banks (logically or physically partitioned), (c) a specialized controller, and (d) a local bus providing signal coupling compatible with the PCI industry standards (this includes but is not limited to PCI-Express, PCI-X, USB 2.0, or functionally similar technologies). The controller handles most of the primitive operations needed to set up and control GPU computation. As a result, the computer's central processing unit (CPU) is freed from this function and is dedicated to other tasks. In this case a few controls (simulation start and stop signals from the CPU and the simulation completion signal back to CPU), GPU programs and input/output data are the information exchanged between CPU and the expansion card. Moreover, since on every time step of the simulation the results from the previous time step are used but not changed, the results are preferably transferred back to CPU in parallel with the computation.

US Patent:

20220383115, Dec 1, 2022

Filed:

Aug 8, 2022

Appl. No.:

17/818015

Inventors:

- Boston MA, US
Liam Debeasi - Brookline MA, US
Heather Ames Versace - Milton MA, US
Jeremy Wurbs - Worcester MA, US
Massimiliano Versace - Milton MA, US
Warren Katz - Cambridge MA, US

Assignee:

Neurala, Inc. - Boston MA

International Classification:

G06N 3/08

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.

US Patent:

20230011901, Jan 12, 2023

Filed:

Jul 11, 2022

Appl. No.:

17/811779

Inventors:

- Boston MA, US
Carly Franca - Boston MA, US
Graham Voysey - Brighton MA, US
Massimiliano Versace - Milton MA, US
Santiago OLIVERA - Brookline MA, US
Vesa TORMANEN - Roslindale MA, US
Alireza Majidi - Medford MA, US
Yiannis Papadopoulos - Waltham MA, US

Assignee:

Neurala, Inc. - Boston MA

International Classification:

G05B 19/418

Abstract:

Industrial quality control is challenging for artificial neural networks (ANNs) and deep neural networks (DNNs) because of the nature of the processed data: there is an abundance of consistent data representing good products, but little data representing bad products. In quality control, the task is changed from conventional DNN task of “recognize what I learned best” to “recognize what I have never seen before.” Lifelong DNN (L-DNN) technology is a hybrid semi-supervised neural architecture that combines the ability of DNNs to be trained, with high precision, on known classes, while being sensitive to any number of unknown classes or class variations. When used for industrial inspection, L-DNN exploits its ability to learn with little and highly unbalanced data. L-DNN's real-time learning capability takes advantage of rare cases of poor-quality products that L-DNN encounters after deployment. L-DNN can be applied to industrial inspections and manufacturing quality control.

US Patent:

20200012943, Jan 9, 2020

Filed:

Sep 17, 2019

Appl. No.:

16/572808

Inventors:

- Boston MA, US
Liam Debeasi - Brookline MA, US
Heather Ames Versace - Milton MA, US
Jeremy Wurbs - Worcester MA, US
Massimiliano Versace - Milton MA, US
Warren Katz - Cambridge MA, US

International Classification:

G06N 3/08
G06K 9/62
G06F 16/58
G06F 16/2455
G06F 17/15

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.

US Patent:

20190240840, Aug 8, 2019

Filed:

Apr 5, 2019

Appl. No.:

16/376109

Inventors:

- Boston MA, US
Massimiliano Versace - Milton MA, US
Tim Barnes - Seattle WA, US

International Classification:

B25J 9/16
G06N 3/00
G06N 3/02
G06K 9/46
G06F 16/22
G06N 5/02
G06F 16/29

Abstract:

The systems and methods disclosed herein include a path integration system that calculates optic flow, infers angular velocity from the flow field, and incorporates this velocity estimate into heading calculations. The resulting system fuses heading estimates from accelerometers, 5 gyroscopes, engine torques, and optic flow to determine self-localization. The system also includes a motivational system that implements a reward drive, both positive and negative, into the system. In some implementations, the drives can include: a) a curiosity drive that encourages exploration of new areas, b) a resource drive that attracts the agent towards the recharging base when the battery is low, and c) a mineral reward drive that attracts the agent 10 towards previously explored scientific targets.

US Patent:

20180330238, Nov 15, 2018

Filed:

May 9, 2018

Appl. No.:

15/975280

Inventors:

- Boston MA, US
Santiago OLIVERA - Brookline MA, US
Jeremy WURBS - Worcester MA, US
Heather Marie AMES - Milton MA, US
Massimiliano VERSACE - Milton MA, US

International Classification:

G06N 3/08
G06N 3/04
G06K 9/62

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.

US Patent:

20160198000, Jul 7, 2016

Filed:

Nov 20, 2015

Appl. No.:

14/947337

Inventors:

- Boston MA, US
Massimiliano Versace - South Boston MA, US
Heather Ames Versace - South Boston MA, US
Gennady Livitz - Belmont MA, US

International Classification:

H04L 29/08
H04L 12/26

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.

US Patent:

20160082597, Mar 24, 2016

Filed:

Nov 20, 2015

Appl. No.:

14/947516

Inventors:

- Boston MA, US
Massimiliano Versace - South Boston MA, US
Tim Barnes - Seattle WA, US

International Classification:

B25J 9/16
G06F 17/30

Abstract:

The systems and methods disclosed herein include a path integration system that calculates optic flow, infers angular velocity from the flow field, and incorporates this velocity estimate into heading calculations. The resulting system fuses heading estimates from accelerometers, gyroscopes, engine torques, and optic flow to determine self-localization. The system also includes a motivational system that implements a reward drive, both positive and negative, into the system. In some implementations, the drives can include: a) a curiosity drive that encourages exploration of new areas, b) a resource drive that attracts the agent towards the recharging base when the battery is low, and c) a mineral reward drive that attracts the agent towards previously explored scientific targets.