arXiv subjects

ShiNung Ching

Publications and source records attributed to ShiNung Ching.

22 records · Page 2Linked to original sources

arXiv2015

Analysis of Recurrent Linear Networks for Enabling Compressed Sensing of Time-Varying Signals

Recent interest has developed around the problem of dynamic compressed sensing, or the recovery of time-varying, sparse signals from limited observations. In this paper, we study how the dynamics of recurrent networks, formulated as general dynamical systems, mediate the recoverability of such signals. We specifically consider the problem of recovering a high-dimensional network input, over time, from observation of only a subset of the network states (i.e., the network output). Our goal is to ascertain how the network dynamics lead to performance advantages, particularly in scenarios where both the input and output are corrupted by disturbance and noise, respectively. For this scenario, we develop bounds on the recovery performance in terms of the dynamics. Conditions for exact recovery in the absence of noise are also formulated. Through several examples, we use the results to highlight how different network characteristics may trade off toward enabling dynamic compressed sensing and how such tradeoffs may manifest naturally in certain classes of neuronal networks.

math.OC↗

arXiv2014

Input Novelty as a Control Metric for Time Varying Linear Systems

This paper introduces a framework for quantitative characterization of the controllability of time-varying linear systems (or networks) in terms of input novelty. The motivation for such an approach comes from the study of biophysical sensory networks in the brain, wherein responsiveness to both energy and salience (or novelty) are presumably critical for mediating behavior and function. Here, we use an inner product to define the angular separation of the current input with respect to the past input history. Then, by constraining input energy, we define a non-convex optimal control problem to obtain the minimally novel input that effects a given state transfer. We provide analytical conditions for existence and uniqueness in continuous-time, as well as an explicit closed-form expression for the solution. In discrete time, we show that a relaxed convex optimization formulation provides the global optimal solution of the original non-convex problem. Finally, we show how the minimum novelty control can be used as a metric to study control properties of large scale recurrent neuronal networks and other complex linear systems. In particular, we highlight unique aspects of a system's controllability that are captured through the novelty-based metric. The result suggests that a multifaceted approach, combining energy-based analysis with other specific metrics, may be useful for obtaining more complete controllability characterizations.

math.OC↗

arXiv2014

On the output of nonlinear systems excited by discrete prolate spheroidal sequences

The discrete prolate spheroidal sequences (DPSSs) - a set of optimally bandlimited sequences with unique properties - are important to applications in both science and engineering. In this work, properties of nonlinear system response due to DPSS excitation are reported. In particular, this output is shown to be approximately orthogonal after passing through a nonlinear, multiple-input multiple-output system with memory under quite general conditions. This work quantifies these conditions in terms of constraints upon the higher-order generalized transfer functions characterizing the Volterra expansion of a MIMO system, the Volterra order of the system, and the DPSS bandwidth parameter W and time-bandwidth parameter NW. The approximate system output orthogonality allows multiple input, multiple-output parameter identification of edge structure in interconnected nonlinear systems using simultaneous, DPSS excitation. This narrowband method of system identification is particularly appealing when compared to classical broadband system excitation in sensitive, neural engineering applications involving electrical stimulation of multiple brain regions. Through the comparison of inner-product and kernel-based narrowband detectors, the utility of this work is demonstrated when identifying narrowband system response of a third-order Volterra system from noisy observations.

math.OC↗

arXiv2014

Controlling Linear Networks with Minimally Novel Inputs

In this paper, we propose a novelty-based metric for quantitative characterization of the controllability of complex networks. This inherently bounded metric describes the average angular separation of an input with respect to the past input history. We use this metric to find the minimally novel input that drives a linear network to a desired state using unit average energy. Specifically, the minimally novel input is defined as the solution of a continuous time, non-convex optimal control problem based on the introduced metric. We provide conditions for existence and uniqueness, and an explicit, closed-form expression for the solution. We support our theoretical results by characterizing the minimally novel inputs for an example of a recurrent neuronal network.

math.OC↗