"CBM (CAM-BRAIN MACHINE)" A Hardware Tool which Evolves a Neural Net Module in a Fraction of a Second and Runs a Million Neuron Artificial Brain in Real Time

نویسندگان

  • Michael KORKIN
  • Hugo de GARIS
  • Felix GERS
  • Hitoshi HEMMI
چکیده

This paper describes a hardware architecture capable of evolving thousands of neural network modules in a matter of minutes and running a simulation of a million neuron modular artificial brain in real time. This new hardware system (called a "CBM", i.e. a CAM-Brain Machine) is an essential component in ATR's ongoing "CAMBrain Project", which aims to build/grow/evolve an artificial brain consisting of a billion neurons by the year 2001. The CBM evolves modules containing some 4000 3D cellular automata (CA) cells with about 100 neurons each, using Xilinx's XC6264 FPGA (Field Programmable Gate Array) chips. This hardware implementation updates two FPGA-based neural modules of 4096 cells each, in parallel, at a 12.8 MHz clock rate, or 105 BILLION CELLS A SECOND. In 1998, this incredible speed will enable the creation of 10,000 evolved modules, which can be assembled into an artificial brain and then run in real time, i.e. the CBM can be used to update the whole CA space which contains all these modules with a total of one million neurons at 25 updates a second (with 100 cycles per module), which is sufficient to directly control robot devices in real time. The CBM is already contracted to be built by the end of 1997. The authors believe that a CBM will make brain building practical.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

The CAM - Brain Machine ( CBM ) An FPGA Based Tool which Evolves Neural Net Circuit Modules in Seconds for Building a 75 Million Neuron Arti cial Brain

This article introduces the \CAM-Brain Machine" (CBM), an FPGA based piece of hardware which implements a genetic algorithm (GA) to evolve a cellular automata (CA) based neural network circuit module, of approximately 1,000 neurons, in about a second, i.e. a complete run of a GA, with 10,000s of circuit growths and performance evaluations. Up to 65,000 of these modules, each of which is evolved...

متن کامل

Atr's "cam-brain Machine" (cbm) and Artiicial Brains an Fpga Based Hardware Tool Which Evolves a Neural Net Circuit Module in a Second and Updates a 40 Million Neuron Artiicial Brain in Real Time

This article introduces ATR's "CAM-Brain Machine" (CBM), an FPGA based piece of hardware which implements a genetic algorithm (GA) to evolve a cellular automata (CA) based neural network circuit module (of approximately 1000 neurons) in about a second (i.e. a complete run of a GA, with 10,000s of circuit growths and performance evaluations). Up to 32000 of these modules (each of which is evolve...

متن کامل

The CAM-Brain Machine (CBM): Real Time Evolution and Update of a 75 Million Neuron FPGA-Based Artificial Brain

This article introduces ATR's \CAM-Brain Machine" (CBM), an FPGA based piece of hardware which implements a genetic algorithm (GA) to evolve a cellular automata (CA) based neural network circuit module, of approximately 1,000 neurons, in about a second, i.e. a complete run of a GA, with 10,000s of circuit growths and performance evaluations. Up to 65,000 of these modules, each of which is evolv...

متن کامل

Evolving Detectors of 2d Patterns on a Simulated Cam-brain Machine, an Evolvable Hardware Tool for Building a 75 Million Neuron Artiicial Brain

This paper presents some simulation results of the evolution of 2D visual pattern recognizers to be implemented very shortly on real hardware, namely the \CAM-Brain Machine" (CBM), an FPGA based piece of evolvable hardware which implements a genetic algorithm (GA) to evolve a 3D cellular automata (CA) based neural network circuit module , of approximately 1,000 neurons, in about a second, i.e. ...

متن کامل

ATR's Artificial Brain (CAM-Brain) Project: A Sample of What Individual CoDi-1Bit Model Evolved Neural Net Modules Can Do

This paper presents a sample of what evolved neural net circuit modules using the socalled ”CoDi-1Bit” neural net model can do. This work is part of an 8 year research project at ATR which aims to build an artificial brain containing a billion neurons by the year 2001, that will be used to control the behaviors of a kitten robot ”Robokoneko”. It looks as though the figure is more likely to be 4...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2007