Con esta herramienta te facilitamos un acceso a todas las ofertas y demandas de tecnología europeas y a búsquedas de socios para participar en propuestas europeas de I+D publicadas en la red Enterprise Europe Network, pudiendo filtrar los resultados para facilitar las búsquedas más acordes con tus necesidades.

¿Quieres recibir estos listados de oportunidades de colaboración en tu correo de forma periódica y personalizada? Date de alta en nuestro Boletín

Los términos de búsqueda han de ser en inglés.

Implementación mediante hardware de grandes redes neuronales artificiales para reconocimiento de sonidos, voz o imágenes

Resumen

Tipo:
Oferta Tecnológica
Referencia:
TOES20190618002
Publicado:
23/07/2019
Caducidad:
23/07/2020
Resumen:
Un grupo de investigación de una universidad española especializado en redes neuronales, minería de datos y aprendizaje automático ha desarrollado un conjunto de técnicas avanzadas para la implementación mediante hardware de diferentes metodologías de aprendizaje automático aplicadas en procesamiento de series de tiempo (reconocimiento de sonidos, voz e imágenes) o predicción de series de tiempo. Gracias al uso de diferentes técnicas de computación aproximada (por ejemplo, computación estocástica) en la implementación de redes neuronales, se reducen considerablemente los requisitos de hardware en comparación con la lógica binaria tradicional. El grupo de investigación busca socios con el fin de establecer acuerdos de cooperación técnica e implementar la tecnología propuesta en las áreas de finanzas, seguridad, etc.

Details

Tittle:
A hardware implementation of large artificial neural networks offered for sounds, speech or image recognition
Summary:
A Spanish university research group, specialized in the field of neural networks, data mining and machine learning, has developed a set of advanced techniques for the hardware implementation of different machine learning methodologies applied for time series processing (sounds, speech and image recognition) or time-series forecasting. The proposed technology implies very important hardware and time reduction ratios. A technical cooperation agreement is being sought.
Description:
Nowadays, internet of things (IoT) and portable devices require the necessity to implement fast and low-power machine learning (ML) techniques in a single integrated circuit (IC) since both power dissipation and process latency are drastically reduced when only high-level information is transmitted from nodes to servers instead of basic raw data. From the different machine learning techniques, artificial neural networks (ANNs) has gained a lot of attention from industry mainly due to the accuracy obtained when solving real-life problems such as image or sound recognition, which normally implies the processing of huge amounts of information. Therefore, there has been a growing interest from the microelectronics industry to implement different artificial intelligence (AI) processes in a single IC.
A Spanish university research group, working in the area neural networks, proposes a new methodology for the hardware implementation of large artificial neural networks based on probabilistic laws. The research group uses advanced techniques based on the use of stochastic computing concepts that allow important hardware reductions as complex computations may be reproduced by a reduced set of digital gates compared to traditional binary logic. With these techniques, large neural networks can be reproduced by the automatic generation of its hardware description code (normally VHDL). Along with the VHDL source codes, detailed technical documentation and training software will be provided for its proper use. Previously, field-programmable gate array (FPGA) implementation demos can be delivered to the potential customers for their evaluation. Those machine learning techniques can be used for both regression and pattern recognition.
A technical cooperation agreement is desired for the implementation of the proposed technology in the areas of finance, security or others.
Advantages and Innovations:
Thanks to the use of different techniques of approximate computing (ex. stochastic computing) in the implementation of the neural network a considerable reduction of the hardware requirements is achieved when compared to traditional binary logic. While area and energy requirements are reduced, accuracy is maintained. The neural network is implemented using a fully digital solution.
Stage of Development:
Available for demonstration
IPs:
Secret Know-how

Partner sought

Type and Role of Partner Sought:
Companies specialized in investment, security, or others, interested in implementing Artificial Neural Networks for forecasting, sound, speech or image recognition.

Client

Type and Size of Client:
R&D Institution
Already Engaged in Trans-National Cooperation:
No
Languages Spoken:
English
Spanish

Keywords

Technology Keywords:
01001002 Sistemas digitales, representación digital