Article

Generating Coherent Patterns of Activity from Chaotic Neural Networks

Department of Neuroscience, Department of Physiology and Cellular Biophysics, Columbia University College of Physicians and Surgeons, New York, NY 10032-2695, USA.
Neuron (Impact Factor: 15.05). 09/2009; 63(4):544-57. DOI: 10.1016/j.neuron.2009.07.018
Source: PubMed

ABSTRACT

Neural circuits display complex activity patterns both spontaneously and when responding to a stimulus or generating a motor output. How are these two forms of activity related? We develop a procedure called FORCE learning for modifying synaptic strengths either external to or within a model neural network to change chaotic spontaneous activity into a wide variety of desired activity patterns. FORCE learning works even though the networks we train are spontaneously chaotic and we leave feedback loops intact and unclamped during learning. Using this approach, we construct networks that produce a wide variety of complex output patterns, input-output transformations that require memory, multiple outputs that can be switched by control inputs, and motor patterns matching human motion capture data. Our results reproduce data on premovement activity in motor and premotor cortex, and suggest that synaptic plasticity may be a more rapid and powerful modulator of network activity than generally appreciated.

Download full-text

Full-text

Available from: David Sussillo, Mar 28, 2014
  • Source
    • "With the development of neural simulators, neural modeling contributes to the advance in computer science research including the fields of artificial intelligence, computer vision, robotics, and data mining. The computational capability of brain-style computing is actively being investigated by several projects (Eliasmith et al., 2012; Furber et al., 2014) and new algorithms inspired by the principle of neural computation are being developed (Gütig and Sompolinsky, 2006; Sussillo and Abbott, 2009; Schmidhuber, 2014). "
    [Show abstract] [Hide abstract]
    ABSTRACT: NeuroFlow is a scalable spiking neural network simulation platform for off-the-shelf high performance computing systems using customizable hardware processors such as Field-Programmable Gate Arrays (FPGAs). Unlike multi-core processors and application-specific integrated circuits, the processor architecture of NeuroFlow can be redesigned and reconfigured to suit a particular simulation to deliver optimized performance, such as the degree of parallelism to employ. The compilation process supports using PyNN, a simulator-independent neural network description language, to configure the processor. NeuroFlow supports a number of commonly used current or conductance based neuronal models such as integrate-and-fire and Izhikevich models, and the spike-timing-dependent plasticity (STDP) rule for learning. A 6-FPGA system can simulate a network of up to ~600,000 neurons and can achieve a real-time performance of 400,000 neurons. Using one FPGA, NeuroFlow delivers a speedup of up to 33.6 times the speed of an 8-core processor, or 2.83 times the speed of GPU-based platforms. With high flexibility and throughput, NeuroFlow provides a viable environment for large-scale neural network simulation.
    Preview · Article · Jan 2016 · Frontiers in Neuroscience
  • Source
    • "The details of the intrinsic organization of the basal ganglia module are skipped in the picture (see Fig. 2) to the cortex feed the fine-scale dynamics of the reservoir, the projections to the basal ganglia feed the differential accumulation of evidence within the channels so as to bias selection on the basis of large-timescale information. Simple static control signals (for instance steady excitatory signals coming from other cortices) would suffice for a reservoir network to be modulated in a similar way as done here by the basal ganglia (Sussillo and Abbott 2009). "
    [Show abstract] [Hide abstract]
    ABSTRACT: The basal ganglia and cortex are strongly implicated in the control of motor preparation and execution. Re-entrant loops between these two brain areas are thought to determine the selection of motor repertoires for instrumental action. The nature of neural encoding and processing in the motor cortex as well as the way in which selection by the basal ganglia acts on them is currently debated. The classic view of the motor cortex implementing a direct mapping of information from perception to muscular responses is challenged by proposals viewing it as a set of dynamical systems controlling muscles. Consequently, the common idea that a competition between relatively segregated cortico-striato-nigro-thalamo-cortical channels selects patterns of activity in the motor cortex is no more sufficient to explain how action selection works. Here, we contribute to develop the dynamical view of the basal ganglia–cortical system by proposing a computational model in which a thalamo-cortical dynamical neural reservoir is modulated by disinhibitory selection of the basal ganglia guided by top-down information, so that it responds with different dynamics to the same bottom-up input. The model shows how different motor trajectories can so be produced by controlling the same set of joint actuators. Furthermore, the model shows how the basal ganglia might modulate cortical dynamics by preserving coarse-grained spatiotemporal information throughout cortico-cortical pathways. Electronic supplementary material The online version of this article (doi:10.1007/s00422-015-0662-6) contains supplementary material, which is available to authorized users.
    Full-text · Article · Nov 2015 · Biological Cybernetics
  • Source
    • "Here we used a modified version of the original recursive least squares ( RLS ) algorithm ( Simon , 2002 ; Jaeger and Haas , 2004 ) based on the FORCE learning formulation ( Sussillo and Abbott , 2009 ) , in order to learn the reservoir - to - readout connection weights W out at each time step , while the CPG input u ( t ) is being fed into the reservoir . The readout weights W out are calculated such that the overall error at the readout neurons is minimized ; thereby the network can learn to accurately transform the CTr - motor signal to the expected foot contact signal , for each walking gait . "
    [Show abstract] [Hide abstract]
    ABSTRACT: Walking animals, like stick insects, cockroaches or ants, demonstrate a fascinating range of locomotive abilities and complex behaviors. The locomotive behaviors can consist of a variety of walking patterns along with adaptation that allow the animals to deal with changes in environmental conditions, like uneven terrains, gaps, obstacles etc. Biological study has revealed that such complex behaviors are a result of a combination of biomechanics and neural mechanism thus representing the true nature of embodied interactions. While the biomechanics helps maintain flexibility and sustain a variety of movements, the neural mechanisms generate movements while making appropriate predictions crucial for achieving adaptation. Such predictions or planning ahead can be achieved by way of internal models that are grounded in the overall behavior of the animal. Inspired by these findings, we present here, an artificial bio-inspired walking system which effectively combines biomechanics (in terms of the body and leg structures) with the underlying neural mechanisms. The neural mechanisms consist of (1) central pattern generator based control for generating basic rhythmic patterns and coordinated movements, (2) distributed (at each leg) recurrent neural network based adaptive forward models with efference copies as internal models for sensory predictions and instantaneous state estimations, and (3) searching and elevation control for adapting the movement of an individual leg to deal with different environmental conditions. Using simulations we show that this bio-inspired approach with adaptive internal models allows the walking robot to perform complex locomotive behaviors as observed in insects, including walking on undulated terrains, crossing large gaps, leg damage adaptations, as well as climbing over high obstacles. Furthermore, we demonstrate that the newly developed recurrent network based approach to online forward models outperforms the adaptive neuron forward models, which have hitherto been the state of the art, to model a subset of similar walking behaviors in walking robots.
    Full-text · Article · Oct 2015 · Frontiers in Neurorobotics
Show more