Search PubMed⌕ Search

Biomedical subjects

Wolfgang Maass

Publications and source records attributed to Wolfgang Maass.

11 recordsLinked to original sources

A statistical analysis of information-processing properties of lamina-specific cortical microcircuit models.

A major challenge for computational neuroscience is to understand the computational function of lamina-specific synaptic connection patterns in stereotypical cortical microcircuits. Previous work on this problem had focused on hypothesized specific computational roles of individual layers and connections between layers and had tested these hypotheses through simulations of abstract neural network models. We approach this problem by studying instead the dynamical system defined by more realistic cortical microcircuit models as a whole and by investigating the influence that its laminar structure has on the transmission and fusion of information within this dynamical system. The circuit models that we examine consist of Hodgkin-Huxley neurons with dynamic synapses, based on detailed data from Thomson and others (2002), Markram and others (1998), and Gupta and others (2000). We investigate to what extent this cortical microcircuit template supports the accumulation and fusion of information contained in generic spike inputs into layer 4 and layers 2/3 and how well it makes this information accessible to projection neurons in layers 2/3 and layer 5. We exhibit specific computational advantages of such data-based lamina-specific cortical microcircuit model by comparing its performance with various types of control models that have the same components and the same global statistics of neurons and synaptic connections but are missing the lamina-specific structure of real cortical microcircuits. We conclude that computer simulations of detailed lamina-specific cortical microcircuit models provide new insight into computational consequences of anatomical and physiological data.

Algorithms↗

On the computational power of threshold circuits with sparse activity.

Circuits composed of threshold gates (McCulloch-Pitts neurons, or perceptrons) are simplified models of neural circuits with the advantage that they are theoretically more tractable than their biological counterparts. However, when such threshold circuits are designed to perform a specific computational task, they usually differ in one important respect from computations in the brain: they require very high activity. On average every second threshold gate fires (sets a 1 as output) during a computation. By contrast, the activity of neurons in the brain is much sparser, with only about 1% of neurons firing. This mismatch between threshold and neuronal circuits is due to the particular complexity measures (circuit size and circuit depth) that have been minimized in previous threshold circuit constructions. In this letter, we investigate a new complexity measure for threshold circuits, energy complexity, whose minimization yields computations with sparse activity. We prove that all computations by threshold circuits of polynomial size with entropy O(log n) can be restructured so that their energy complexity is reduced to a level near the entropy of circuit states. This entropy of circuit states is a novel circuit complexity measure, which is of interest not only in the context of threshold circuits but for circuit complexity in general. As an example of how this measure can be applied, we show that any polynomial size threshold circuit with entropy O(log n) can be simulated by a polynomial size threshold circuit of depth 3. Our results demonstrate that the structure of circuits that result from a minimization of their energy complexity is quite different from the structure that results from a minimization of previously considered complexity measures, and potentially closer to the structure of neural circuits in the nervous system. In particular, different pathways are activated in these circuits for different classes of inputs. This letter shows that such circuits with sparse activity have a surprisingly large computational power.

Brain↗

Fading memory and kernel properties of generic cortical microcircuit models.

It is quite difficult to construct circuits of spiking neurons that can carry out complex computational tasks. On the other hand even randomly connected circuits of spiking neurons can in principle be used for complex computational tasks such as time-warp invariant speech recognition. This is possible because such circuits have an inherent tendency to integrate incoming information in such a way that simple linear readouts can be trained to transform the current circuit activity into the target output for a very large number of computational tasks. Consequently we propose to analyze circuits of spiking neurons in terms of their roles as analog fading memory and non-linear kernels, rather than as implementations of specific computational operations and algorithms. This article is a sequel to [W. Maass, T. Natschläger, H. Markram, Real-time computing without stable states: a new framework for neural computation based on perturbations, Neural Comput. 14 (11) (2002) 2531-2560, Online available as #130 from: ], and contains new results about the performance of generic neural microcircuit models for the recognition of speech that is subject to linear and non-linear time-warps, as well as for computations on time-varying firing rates. These computations rely, apart from general properties of generic neural microcircuit models, just on capabilities of simple linear readouts trained by linear regression. This article also provides detailed data on the fading memory property of generic neural microcircuit models, and a quick review of other new results on the computational power of such circuits of spiking neurons.

Action Potentials↗

Dynamics of information and emergent computation in generic neural microcircuit models.

Numerous methods have already been developed to estimate the information contained in single spike trains. In this article we explore efficient methods for estimating the information contained in the simultaneous firing activity of hundreds of neurons. Obviously such methods are needed to analyze data from multi-unit recordings. We test these methods on generic neural microcircuit models consisting of 800 neurons, and analyze the temporal dynamics of information about preceding spike inputs in such circuits. It turns out that information spreads with high speed in such generic neural microcircuit models, thereby supporting-without the postulation of any additional neural or synaptic mechanisms--the possibility of ultra-rapid computations on the first input spikes.

Algorithms↗

A model for the interaction of oscillations and pattern generation with real-time computing in generic neural microcircuit models.

It is shown that real-time computations on spike patterns and temporal integration of information in neural microcircuit models are compatible with potentially descruptive additional inputs such as oscillations. A minor change in the connection statistics of such circuits (making synaptic connections to more distal target neurons more likely for excitatory than for inhibitory neurons) endows such generic neural microcircuit model with the ability to generate periodic patterns autonomously. We show that such pattern generation can also be multiplexed with pattern classification and temporal integration of information in the same neural circuit. These results can be interpreted as showing that periodic activity provides a second channel for communication in neural systems which can be used to synchronize or coordinate spatially separated processes, without encumbering local real-time computations on spike trains in diverse neural circuits.

Action Potentials↗

Movement generation with circuits of spiking neurons.

How can complex movements that take hundreds of milliseconds be generated by stereotypical neural microcircuits consisting of spiking neurons with a much faster dynamics? We show that linear readouts from generic neural microcircuit models can be trained to generate basic arm movements. Such movement generation is independent of the arm model used and the type of feedback that the circuit receives. We demonstrate this by considering two different models of a two-jointed arm, a standard model from robotics and a standard model from biology, that each generates different kinds of feedback. Feedback that arrives with biologically realistic delays of 50 to 280 ms turns out to give rise to the best performance. If a feedback with such desirable delay is not available, the neural microcircuit model also achieves good performance if it uses internally generated estimates of such feedback. Existing methods for movement generation in robotics that take the particular dynamics of sensors and actuators into account (embodiment of motor systems) are taken one step further with this approach, which provides methods for also using the embodiment of motion generation circuitry, that is, the inherent dynamics and spatial structure of neural circuits, for the generation of movement.

Action Potentials↗

What can a neuron learn with spike-timing-dependent plasticity?

Spiking neurons are very flexible computational modules, which can implement with different values of their adjustable synaptic parameters an enormous variety of different transformations F from input spike trains to output spike trains. We examine in this letter the question to what extent a spiking neuron with biologically realistic models for dynamic synapses can be taught via spike-timing-dependent plasticity (STDP) to implement a given transformation F. We consider a supervised learning paradigm where during training, the output of the neuron is clamped to the target signal (teacher forcing). The well-known perceptron convergence theorem asserts the convergence of a simple supervised learning algorithm for drastically simplified neuron models (McCulloch-Pitts neurons). We show that in contrast to the perceptron convergence theorem, no theoretical guarantee can be given for the convergence of STDP with teacher forcing that holds for arbitrary input spike patterns. On the other hand, we prove that average case versions of the perceptron convergence theorem hold for STDP in the case of uncorrelated and correlated Poisson input spike trains and simple models for spiking neurons. For a wide class of cross-correlation functions of the input spike trains, the resulting necessary and sufficient condition can be formulated in terms of linear separability, analogously as the well-known condition of learnability by perceptrons. However, the linear separability criterion has to be applied here to the columns of the correlation matrix of the Poisson input. We demonstrate through extensive computer simulations that the theoretically predicted convergence of STDP with teacher forcing also holds for more realistic models for neurons, dynamic synapses, and more general input distributions. In addition, we show through computer simulations that these positive learning results hold not only for the common interpretation of STDP, where STDP changes the weights of synapses, but also for a more realistic interpretation suggested by experimental data where STDP modulates the initial release probability of dynamic synapses.

Action Potentials↗

Input prediction and autonomous movement analysis in recurrent circuits of spiking neurons.

Temporal integration of information and prediction of future sensory inputs are assumed to be important computational tasks of generic cortical microcircuits. It has remained open how cortical microcircuits could possibly achieve this, especially since they consist--in contrast to most neural network models--of neurons and synapses with heterogeneous dynamic responses. It turns out, however, that the diversity of computational units increases the capability of microcircuit models for temporal integration. Furthermore the prediction of future input may be rather easy for such circuits since it suffices to train the readouts from such microcircuits. In this article we show that very simple readouts from a generic recurrently connected circuit of integrate-and-fire neurons with diverse dynamic synapses can be trained in an unsupervised manner to predict movements of different objects, that move within an unlimited number of combinations of speed, angle, and offset over a simulated sensory field. The autonomously trained microcircuit model is also able to compute the direction of motion, which is a computationally difficult problem ('aperture problem') since it requires disambiguation of local sensory readings through the context of other sensory readings at the current and preceding moments. Furthermore the same circuit can be trained simultaneously in a supervised manner to also report the shape and velocity of the moving object. Finally it is shown that the trained neural circuit supports novelty detection and the generation of 'imagined movements'. Altogether the results of this article suggest that it is not necessary to construct specific and biologically unrealistic neural circuit models for specific sensory processing tasks, since 'found' generic cortical microcircuit models in combination with very simple perceptron-like readouts can easily be trained to solve such computational tasks.

Action Potentials↗

Synapses as dynamic memory buffers.

This article throws new light on the possible role of synapses in information transmission through theoretical analysis and computer simulations. We show that the internal dynamic state of a synapse may serve as a transient memory buffer that stores information about the most recent segment of the spike train that was previously sent to this synapse. This information is transmitted to the postsynaptic neuron through the amplitudes of the postsynaptic response for the next few spikes. In fact, we show that most of this information about the preceding spike train is already contained in the postsynaptic response for just two additional spikes. It is demonstrated that the postsynaptic neuron receives simultaneously information about the specific type of synapse which has transmitted these pulses. In view of recent findings by Gupta et al. [Science, 287 (2000) 273] that different types of synapses are characteristic for specific types of presynaptic neurons, the postsynaptic neuron receives in this way partial knowledge about the identity of the presynaptic neuron from which it has received information. Our simulations are based on recent data about the dynamics of GABAergic synapses. We show that the relatively large number of synaptic release sites that make up a GABAergic synaptic connection makes these connections suitable for such complex information transmission processes.

Computer Simulation↗

Real-time computing without stable states: a new framework for neural computation based on perturbations.

A key challenge for neural modeling is to explain how a continuous stream of multimodal input from a rapidly changing environment can be processed by stereotypical recurrent circuits of integrate-and-fire neurons in real time. We propose a new computational model for real-time computing on time-varying input that provides an alternative to paradigms based on Turing machines or attractor neural networks. It does not require a task-dependent construction of neural circuits. Instead, it is based on principles of high-dimensional dynamical systems in combination with statistical learning theory and can be implemented on generic evolved or found recurrent circuitry. It is shown that the inherent transient dynamics of the high-dimensional dynamical system formed by a sufficiently large and heterogeneous neural circuit may serve as universal analog fading memory. Readout neurons can learn to extract in real time from the current state of such recurrent neural circuit information about current and past inputs that may be needed for diverse tasks. Stable internal states are not required for giving a stable output, since transient internal states can be transformed by readout neurons into stable target outputs due to the high dimensionality of the dynamical system. Our approach is based on a rigorous computational model, the liquid state machine, that, unlike Turing machines, does not require sequential transitions between well-defined discrete internal states. It is supported, as the Turing machine is, by rigorous mathematical results that predict universal computational power under idealized conditions, but for the biologically more realistic scenario of real-time processing of time-varying inputs. Our approach provides new perspectives for the interpretation of neural coding, the design of experiments and data analysis in neurophysiology, and the solution of problems in robotics and neurotechnology.

Action Potentials↗