Brain Derived Vision Algorithm on High Performance Architectures

Nageswaran, Jayram Moorkanikara; Felch, Andrew; Chandrasekhar, Ashok; Dutt, Nikil; Granger, Richard; Nicolau, Alex; Veidenbaum, Alex
August 2009
International Journal of Parallel Programming;Aug2009, Vol. 37 Issue 4, p345
Academic Journal
Even though computing systems have increased the number of transistors, the switching speed, and the number of processors, most programs exhibit limited speedup due to the serial dependencies of existing algorithms. Analysis of intrinsically parallel systems such as brain circuitry have led to the identification of novel architecture designs, and also new algorithms than can exploit the features of modern multiprocessor systems. In this article we describe the details of a brain derived vision (BDV) algorithm that is derived from the anatomical structure, and physiological operating principles of thalamo-cortical brain circuits. We show that many characteristics of the BDV algorithm lend themselves to implementation on IBM CELL architecture, and yield impressive speedups that equal or exceed the performance of specialized solutions such as FPGAs. Mapping this algorithm to the IBM CELL is non-trivial, and we suggest various approaches to deal with parallelism, task granularity, communication, and memory locality. We also show that a cluster of three PS3s (or more) containing IBM CELL processors provides a promising platform for brain derived algorithms, exhibiting speedup of more than 140 × over a desktop PC implementation, and thus enabling real-time object recognition for robotic systems.


Related Articles

  • A Parallel Circuit Simulator for Iterative Power Grids Optimization System.  // Circuits & Systems;Apr2012, Vol. 3 Issue 2, p153 

    The article examines a parallel circuit simulator that demonstrates high efficiency to promote iterative power grid optimization. It discusses the use of hardware-oriented fixed point arithmetic and dedicated hardware development adopting parallel processing. The result reveals a simulation with...

  • Self-repair in a bidirectionally coupled astrocyte-neuron (AN) system based on retrograde signaling. Wade, John; McDaid, Liam; Harkin, Jim; Crunelli, Vincenzo; Kelso, Scott // Frontiers in Computational Neuroscience;Sep2012, Vol. 6, p1 

    In this paper we demonstrate that retrograde signaling via astrocytes may underpin self-repair in the brain. Faults manifest themselves in silent or near silent neurons caused by low transmission probability (PR) synapses; the enhancement of the transmission PR of a healthy neighboring synapse...

  • A Solid-State Neuron for Spiking Neural Network Implementation. Yajie Chen; Hall, Steve; McDaid, Liam; Buiu, Octavian; Kelly, Peter // Engineering Letters;2008, Vol. 16 Issue 1, p83 

    This paper presents a compact analog neuron cell incorporating an array of charge-coupled synapses connected via a common output terminal. The novel silicon synapse is based on a two stage charge-coupled device where the weighting functionality can be integrated into the first stage. A...

  • NEUROELECTRONICS: Silicon neurons. López, Juan Carlos // Nature Reviews Neuroscience;Oct2001, Vol. 2 Issue 10, p681 

    Discusses a study which examined the possibility of transmitting signals through a silicon-neuron-neuron-silicon circuit. Formation of electrical synapses with neighboring cells; Limitations to the development of complex and efficient circuits.

  • Pass That CEO a Pocket Protector. Braune, Bernd // Electronic News;7/2/2001, Vol. 47 Issue 27, p35 

    Discusses the implications of the proliferation of silicon content in electronic chip design. Factors to be considered by chief executive officers; Assessment on the design costs for a transistor chip; Changes in semiconductor designs.

  • Physical origin and characteristics of gate capacitance in silicon metal-oxide-semiconductor... Nakajima, Yasuyuki; Horiguchi, Seiji; Shoji, Masanari; Omura, Yasuhisa // Journal of Applied Physics;5/1/1998, Vol. 83 Issue 9, p4788 

    Presents a study to describe a theoretical model of gate capacitance (Cg), while assessing the physical origin and characteristics of Cg in silicon metal-oxide-semiconductor field-effect transistors. Methodology used to conduct study; Indication of findings; Discussion on results.

  • Enhancement of presynaptic neuronal excitability by correlated presynaptic and postsynaptic spiking. Ganguly, Karunesh; Kiss, Laszlo; Poo, Mu-ming // Nature Neuroscience;Oct2000, Vol. 3 Issue 10, p1018 

    Use-dependent modifications, such as long-term potentiation (LTP) of synaptic efficacy, are believed to be essential for information storage in the nervous system. Repetitive correlated spiking of presynaptic and postsynaptic neurons can induce LTP at excitatory glutamatergic synapses. In...

  • IPSP.  // Taber's Cyclopedic Medical Dictionary (2009);2009, Issue 21, p1230 

    A definition of the acronym "IPSP" which refers to inhibitory postsynaptic potential is presented.

  • From the editors. Wiedemann, Claudia; Whalley, Katherine; Welberg, Leonie // Nature Reviews Neuroscience;Dec2007, Vol. 8 Issue 12, p905 

    The article discusses the topics published within the issue, including one on the creation of the Brainbow genetic constructs that allow researchers to visualize individual neurons and another on the role of SUMOylation in neuronal development, synaptic plasticity and neurological disorders.


Read the Article


Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics