Limits...
Criticality as a Set-Point for Adaptive Behavior in Neuromorphic Hardware.

Srinivasa N, Stepp ND, Cruz-Albrecht J - Front Neurosci (2015)

Bottom Line: Experiment, theory and recent models have shown that neuronal networks at criticality demonstrate optimal information transfer, learning and information processing capabilities that affect behavior.In this perspective article, we argue that understanding how large scale neuromorphic electronics can be designed to enable emergent adaptive behavior will require an understanding of how networks emulated by such hardware can self-tune local parameters to maintain criticality as a set-point.We believe that such capability will enable the design of truly scalable intelligent systems using neuromorphic hardware that embrace complexity in network dynamics rather than avoiding it.

View Article: PubMed Central - PubMed

Affiliation: Information and System Sciences Lab, Center for Neural and Emergent Systems, HRL Laboratories LLC Malibu, CA, USA.

ABSTRACT
Neuromorphic hardware are designed by drawing inspiration from biology to overcome limitations of current computer architectures while forging the development of a new class of autonomous systems that can exhibit adaptive behaviors. Several designs in the recent past are capable of emulating large scale networks but avoid complexity in network dynamics by minimizing the number of dynamic variables that are supported and tunable in hardware. We believe that this is due to the lack of a clear understanding of how to design self-tuning complex systems. It has been widely demonstrated that criticality appears to be the default state of the brain and manifests in the form of spontaneous scale-invariant cascades of neural activity. Experiment, theory and recent models have shown that neuronal networks at criticality demonstrate optimal information transfer, learning and information processing capabilities that affect behavior. In this perspective article, we argue that understanding how large scale neuromorphic electronics can be designed to enable emergent adaptive behavior will require an understanding of how networks emulated by such hardware can self-tune local parameters to maintain criticality as a set-point. We believe that such capability will enable the design of truly scalable intelligent systems using neuromorphic hardware that embrace complexity in network dynamics rather than avoiding it.

No MeSH data available.


(A) Tuning process required if synaptic weights or other low-level parameters need to be tuned for specific network inputs. (B) Tuning process for a self-tuning critical network. In this case, the network is tuned at a high level to select for critical dynamics that is adaptively maintained internally.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4664726&req=5

Figure 1: (A) Tuning process required if synaptic weights or other low-level parameters need to be tuned for specific network inputs. (B) Tuning process for a self-tuning critical network. In this case, the network is tuned at a high level to select for critical dynamics that is adaptively maintained internally.

Mentions: To achieve a system that self-tunes toward criticality, i.e., treats criticality as set-point, we performed a parameter search at the level of global STP, STDP, and synaptic kinetics parameters. This search resulted in several network configurations that maintained themselves in a state of criticality, even after being perturbed by external inputs (For details about the parameter search and associated measurements, see Stepp et al., 2015). A parameter search of this sort should be distinguished from a search at the level of individual synaptic characteristics. For instance, tuning a network to maintain useful activity while receiving certain input might involve setting individual synaptic weights. These weights would generally have to be set differently for different classes of input. The search for self-tuning criticality happens at a more global network level, and the resulting network configuration is appropriate for many classes of input. The two different approaches are depicted in Figure 1.


Criticality as a Set-Point for Adaptive Behavior in Neuromorphic Hardware.

Srinivasa N, Stepp ND, Cruz-Albrecht J - Front Neurosci (2015)

(A) Tuning process required if synaptic weights or other low-level parameters need to be tuned for specific network inputs. (B) Tuning process for a self-tuning critical network. In this case, the network is tuned at a high level to select for critical dynamics that is adaptively maintained internally.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4664726&req=5

Figure 1: (A) Tuning process required if synaptic weights or other low-level parameters need to be tuned for specific network inputs. (B) Tuning process for a self-tuning critical network. In this case, the network is tuned at a high level to select for critical dynamics that is adaptively maintained internally.
Mentions: To achieve a system that self-tunes toward criticality, i.e., treats criticality as set-point, we performed a parameter search at the level of global STP, STDP, and synaptic kinetics parameters. This search resulted in several network configurations that maintained themselves in a state of criticality, even after being perturbed by external inputs (For details about the parameter search and associated measurements, see Stepp et al., 2015). A parameter search of this sort should be distinguished from a search at the level of individual synaptic characteristics. For instance, tuning a network to maintain useful activity while receiving certain input might involve setting individual synaptic weights. These weights would generally have to be set differently for different classes of input. The search for self-tuning criticality happens at a more global network level, and the resulting network configuration is appropriate for many classes of input. The two different approaches are depicted in Figure 1.

Bottom Line: Experiment, theory and recent models have shown that neuronal networks at criticality demonstrate optimal information transfer, learning and information processing capabilities that affect behavior.In this perspective article, we argue that understanding how large scale neuromorphic electronics can be designed to enable emergent adaptive behavior will require an understanding of how networks emulated by such hardware can self-tune local parameters to maintain criticality as a set-point.We believe that such capability will enable the design of truly scalable intelligent systems using neuromorphic hardware that embrace complexity in network dynamics rather than avoiding it.

View Article: PubMed Central - PubMed

Affiliation: Information and System Sciences Lab, Center for Neural and Emergent Systems, HRL Laboratories LLC Malibu, CA, USA.

ABSTRACT
Neuromorphic hardware are designed by drawing inspiration from biology to overcome limitations of current computer architectures while forging the development of a new class of autonomous systems that can exhibit adaptive behaviors. Several designs in the recent past are capable of emulating large scale networks but avoid complexity in network dynamics by minimizing the number of dynamic variables that are supported and tunable in hardware. We believe that this is due to the lack of a clear understanding of how to design self-tuning complex systems. It has been widely demonstrated that criticality appears to be the default state of the brain and manifests in the form of spontaneous scale-invariant cascades of neural activity. Experiment, theory and recent models have shown that neuronal networks at criticality demonstrate optimal information transfer, learning and information processing capabilities that affect behavior. In this perspective article, we argue that understanding how large scale neuromorphic electronics can be designed to enable emergent adaptive behavior will require an understanding of how networks emulated by such hardware can self-tune local parameters to maintain criticality as a set-point. We believe that such capability will enable the design of truly scalable intelligent systems using neuromorphic hardware that embrace complexity in network dynamics rather than avoiding it.

No MeSH data available.