Conference record. Asilomar Conference on Signals, Systems & Computers最新文献

In simplified models of neocortical circuits, inhibition is either modeled in a feedforward manner or through mutual inhibitory interactions that provide for competition between neuronal populations. By contrast, recent work has suggested a critical role for recurrent inhibition as a negative feedback element that stabilizes otherwise unstable recurrent excitation. Here, we show how models based upon a motif of recurrently connected "E-I" pairs of excitatory and inhibitory units can be used to describe experimental observations in sensory and memory networks. In a sensory network model of binocular rivalry, a model based on competing E-I motifs captures psychophysical observations about how incongruous images presented to the two eyes compete. In a model of cortical working memory, an architecturally similar model with modified synaptic time constants can mathematically accumulate signals into a working memory buffer in a manner that is robust to the abrupt removal of cells. These results suggest the inhibition-stabilized E-I motif as a fundamental building block for models of a wide array of neocortical dynamics.

Recent studies suggest that disruptions in resting state functional connectivity - a measure of stationary statistical association between brain regions - can be used as an objective marker of brain injury. However, fewer characterizations have examined the disruption of intrinsic brain dynamics after brain injury. Here, we examine this issue using electroencephalographic (EEG) data from brain-injured patients, together with a control analysis wherein we quantify the effect of the injury on the ability of intrinsic event responses to traverse their respective state spaces. More specifically, the lability of intrinsically evoked brain activity was assessed by collapsing three sigma event responses in all channels of the obtained EEG signals into a low-dimensional space. The directional derivative of these responses was then used to assay the extent to which brain activity reaches low-variance subspaces. Our findings suggest that intrinsic dynamics extracted from resting state EEG signals can differentiate various levels of consciousness in severe cases of coma. More specifically the cost of moving from one state to another in the state-space trajectories of the underlying dynamics becomes lower as the level of consciousness of patients deteriorates.

We consider configuration of multiple mobile sensors to explore and refine knowledge in an unknown field. After some initial discovery, it is desired to collect data from the regions that are far away from the current sensor trajectories in order to favor the exploration purposes, while simultaneously, exploring the vicinity of known interesting phenomena to refine the measurements. Since the collected data only provide us with local information, there is no optimal solution to be sought for the next trajectory of sensors. Using Gaussian process regression, we provide a simple framework that accounts for both the conflicting data refinement and exploration goals, and to make reasonable decisions for the trajectories of mobile sensors.

Sleep is important for normal brain function, and sleep disruption is comorbid with many neurological diseases. There is a growing mechanistic understanding of the neurological basis for sleep regulation that is beginning to lead to mechanistic mathematically described models. It is our objective to validate the predictive capacity of such models using data assimilation (DA) methods. If such methods are successful, and the models accurately describe enough of the mechanistic functions of the physical system, then they can be used as sophisticated observation systems to reveal both system changes and sources of dysfunction with neurological diseases and identify routes to intervene. Here we report on extensions to our initial efforts [1] at applying unscented Kalman Filter (UKF) to models of sleep regulation on three fronts: tools for multi-parameter fitting; a sophisticated observation model to apply the UKF for observations of behavioral state; and comparison with data recorded from brainstem cell groups thought to regulate sleep.

Recovery of sparse signals with unknown clustering pattern in the case of having partial erroneous prior knowledge on the supports of the signal is considered. In this case, we provide a modified sparse Bayesian learning model to incorporate prior knowledge and simultaneously learn the unknown clustering pattern. For this purpose, we add one more layer to support-aided sparse Bayesian learning algorithm (SA-SBL). This layer adds a prior on the shape parameters of Gamma distributions, those modeled to account for the precision of the solution elements. We make the shape parameters depend on the total variations on the estimated supports of the solution. Based on the simulation results, we show that the proposed algorithm is able to modify its erroneous prior knowledge on the supports of the solution and learn the clustering pattern of the true signal by filtering out the incorrect supports from the estimated support set.

Neural data analysis has increasingly incorporated causal information to study circuit connectivity. Dimensional reduction forms the basis of most analyses of large multivariate time series. Here, we present a new, multitaper-based decomposition for stochastic, multivariate time series that acts on the covariance of the time series at all lags, C(τ), as opposed to standard methods that decompose the time series, X(t), using only information at zero-lag. In both simulated and neural imaging examples, we demonstrate that methods that neglect the full causal structure may be discarding important dynamical information in a time series.

Previous modeling studies have demonstrated that lateral inhibition contributes to enhanced precision in sensory networks. That is, inhibitory connections reduce the spread of activity and repress neighboring cells, increasing the reliability of a sensory response. However, much less is understood about how connections that spread activity might contribute to the processing of sensory stimuli in the context of a sensory discrimination task. In this work, we examine the role of excitatory connections and gap junctions in network dynamics and some contributions to sensory discrimination.

Six of the ten leading causes of death in the United States can be directly linked to diet. Measuring accurate dietary intake, the process of determining what someone eats is considered to be an open research problem in the nutrition and health fields. We are developing image-based tools in order to automatically obtain accurate estimates of what foods a user consumes. We have developed a novel food record application using the embedded camera in a mobile device. This paper describes the current status of food image analysis and overviews problems that still need to be addressed.

Finding the solution of single measurement vector (SMV) problem with an unknown block-sparsity structure is considered. Here, we propose a sparse Bayesian learning (SBL) algorithm simplified via the approximate message passing (AMP) framework. In order to encourage the block-sparsity structure, we incorporate a parameter called Sigma-Delta as a measure of clumpiness in the supports of the solution. Using the AMP framework reduces the computational load of the proposed SBL algorithm and as a result makes it faster. Furthermore, in terms of the mean-squared error between the true and the reconstructed solution, the algorithm demonstrates an encouraging improvement compared to the other algorithms.