Owing to the proximity of the ear canal to the central nervous system, in-ear electrophysiologica... more Owing to the proximity of the ear canal to the central nervous system, in-ear electrophysiological systems can be used to unobtrusively monitor brain states. Here, by taking advantage of the ear’s exocrine sweat glands, we describe an in-ear integrated array of electrochemical and electrophysiological sensors placed on a flexible substrate surrounding a user-generic earphone for the simultaneous monitoring of lactate concentration and brain states via electroencephalography, electrooculography and electrodermal activity. In volunteers performing an acute bout of exercise, the device detected elevated lactate levels in sweat concurrently with the modulation of brain activity across all electroencephalography frequency bands. Simultaneous and continuous unobtrusive in-ear monitoring of metabolic biomarkers and brain electrophysiology may allow for the discovery of dynamic and synergetic interactions between brain and body biomarkers in real-world settings for long-term health monitori...
Many powerful neural network (NN) models such as probabilistic graphical models (PGMs) and recurr... more Many powerful neural network (NN) models such as probabilistic graphical models (PGMs) and recurrent neural networks (RNNs) require flexibility in dataflow and weight access patterns as shown in Fig. 33.1.1 Typically, Compute-In-Memory (CIM) designs do not implement such dataflows or do so by replicating circuits at the memory periphery such as ADCs/neurons along both the rows and columns of the memory array, leading to an overhead in operation. This paper describes a CIM architecture implemented in a 130nm CMOS/RRAM process, that delivers the highest reported computational energy-efficiency of 74 tera-multiply-accumulates per second per watt (TMACS/W) for RRAM-based CIM architectures while simultaneously offering dataflow reconfigurability to address the limitations of previous designs. This is made possible through two key features: 1) a runtime reconfigurable dataflow with in-situ access to RRAM array and its transpose for efficient access to NN weights and 2) a voltage sensing stochastic integrate-and-fire analog neuron (I&F) that is reused for correlated double sampling (CDS), stochastic voltage integration, and threshold comparison.
We present an efficient and scalable partitioning method for mapping large-scale neural network m... more We present an efficient and scalable partitioning method for mapping large-scale neural network models to reconfigurable neuromorphic hardware. The partitioning framework is optimized for compute-balanced, memory -efficient parallel processing targeting low-latency execution and dense synaptic storage, with minimal routing across various compute cores. We demonstrate highly scalable and efficient partitioning for connectivity-aware and hierarchical address-event routing resource-optimized mapping, significantly reducing the total communication volume recursively when compared to random balanced assignment. We evaluate the partitioning algorithm on synthetic small-world networks with varying degrees of sparsity factor and fan-out. The combination of our method and practical results suggest a promising path towards extending to very large-scale networks and more degrees of hierarchy.
Clinical assessment of the human auditory system is an integral part of evaluating the health of ... more Clinical assessment of the human auditory system is an integral part of evaluating the health of a patient’s cognitive processes. Conventional tests performed by audiologists include the Auditory Steady State Response (ASSR) and the Auditory Brainstem Response (ABR), both of which present an audio stimulus to the patient in order to elicit a change in brain state measurable by electroencephalography (EEG) techniques. Spatial monitoring of the electrophysiological activity in the auditory cortex, temporal cortex, and brain stem during auditory stimulus evaluation can be used to pinpoint to location of auditory dysfunction along the auditory pathway. However, given the obtrusive nature of conventional auditory evaluation techniques and the lack of information about sound transduction and cochlear dynamics usually irrecoverable by EEG, a better approach is needed to improve its clinical utility. Here, we present an in-ear device for auditory health assessment that integrates a sound engine for stimulation and high-density dry-electrode EEG for real-time simultaneous recording of brain activity. This system provides ease-of-use and patient comfort. We also investigate the auditory transfer function of the hearing system as an intricate convolution of the tympanic membrane, middle ear bones, and the cochlear subsystems.
ABSTRACT Biological systems which inspire connectionist architecture are based upon axonal point ... more ABSTRACT Biological systems which inspire connectionist architecture are based upon axonal point to point connections. As we attempt to engineer ever larger analogues of these neural networks we are forced to multiplex neural signals over time shared paths. This can alter the timing of signal arrival which is critical in oscillatory networks and leads to the bandwidth-synchrony dilemma. This paper further describes the problem, especially for analog VLSI systems, and approaches under consideration for the solution
Journal of Consciousness Exploration & Research, 2010
For the reasons discussed herein, it makes sense to treat consciousness as a process pervasive in... more For the reasons discussed herein, it makes sense to treat consciousness as a process pervasive in nature, at all levels of complexity. It can be seen as having a type of self-similarity. Recall that time supervenes on change, change requires contrast, and the contrast has to be detected. Whatever systems are changing are sensing and recording their reaction to the contrast in their behavior and in their state change.
A multi-chip analog neural network system capable of prototyping networks with as many as 81920 s... more A multi-chip analog neural network system capable of prototyping networks with as many as 81920 synaptic connections and 1024 neurons is described. The neural network architecture is reconfigurable by routing all neuron activation values through a host computer which can re-map the network connectivity by changing a look-up table in memory. Once a network is successfully prototyped, it is hardwired
Owing to the proximity of the ear canal to the central nervous system, in-ear electrophysiologica... more Owing to the proximity of the ear canal to the central nervous system, in-ear electrophysiological systems can be used to unobtrusively monitor brain states. Here, by taking advantage of the ear’s exocrine sweat glands, we describe an in-ear integrated array of electrochemical and electrophysiological sensors placed on a flexible substrate surrounding a user-generic earphone for the simultaneous monitoring of lactate concentration and brain states via electroencephalography, electrooculography and electrodermal activity. In volunteers performing an acute bout of exercise, the device detected elevated lactate levels in sweat concurrently with the modulation of brain activity across all electroencephalography frequency bands. Simultaneous and continuous unobtrusive in-ear monitoring of metabolic biomarkers and brain electrophysiology may allow for the discovery of dynamic and synergetic interactions between brain and body biomarkers in real-world settings for long-term health monitori...
Many powerful neural network (NN) models such as probabilistic graphical models (PGMs) and recurr... more Many powerful neural network (NN) models such as probabilistic graphical models (PGMs) and recurrent neural networks (RNNs) require flexibility in dataflow and weight access patterns as shown in Fig. 33.1.1 Typically, Compute-In-Memory (CIM) designs do not implement such dataflows or do so by replicating circuits at the memory periphery such as ADCs/neurons along both the rows and columns of the memory array, leading to an overhead in operation. This paper describes a CIM architecture implemented in a 130nm CMOS/RRAM process, that delivers the highest reported computational energy-efficiency of 74 tera-multiply-accumulates per second per watt (TMACS/W) for RRAM-based CIM architectures while simultaneously offering dataflow reconfigurability to address the limitations of previous designs. This is made possible through two key features: 1) a runtime reconfigurable dataflow with in-situ access to RRAM array and its transpose for efficient access to NN weights and 2) a voltage sensing stochastic integrate-and-fire analog neuron (I&F) that is reused for correlated double sampling (CDS), stochastic voltage integration, and threshold comparison.
We present an efficient and scalable partitioning method for mapping large-scale neural network m... more We present an efficient and scalable partitioning method for mapping large-scale neural network models to reconfigurable neuromorphic hardware. The partitioning framework is optimized for compute-balanced, memory -efficient parallel processing targeting low-latency execution and dense synaptic storage, with minimal routing across various compute cores. We demonstrate highly scalable and efficient partitioning for connectivity-aware and hierarchical address-event routing resource-optimized mapping, significantly reducing the total communication volume recursively when compared to random balanced assignment. We evaluate the partitioning algorithm on synthetic small-world networks with varying degrees of sparsity factor and fan-out. The combination of our method and practical results suggest a promising path towards extending to very large-scale networks and more degrees of hierarchy.
Clinical assessment of the human auditory system is an integral part of evaluating the health of ... more Clinical assessment of the human auditory system is an integral part of evaluating the health of a patient’s cognitive processes. Conventional tests performed by audiologists include the Auditory Steady State Response (ASSR) and the Auditory Brainstem Response (ABR), both of which present an audio stimulus to the patient in order to elicit a change in brain state measurable by electroencephalography (EEG) techniques. Spatial monitoring of the electrophysiological activity in the auditory cortex, temporal cortex, and brain stem during auditory stimulus evaluation can be used to pinpoint to location of auditory dysfunction along the auditory pathway. However, given the obtrusive nature of conventional auditory evaluation techniques and the lack of information about sound transduction and cochlear dynamics usually irrecoverable by EEG, a better approach is needed to improve its clinical utility. Here, we present an in-ear device for auditory health assessment that integrates a sound engine for stimulation and high-density dry-electrode EEG for real-time simultaneous recording of brain activity. This system provides ease-of-use and patient comfort. We also investigate the auditory transfer function of the hearing system as an intricate convolution of the tympanic membrane, middle ear bones, and the cochlear subsystems.
ABSTRACT Biological systems which inspire connectionist architecture are based upon axonal point ... more ABSTRACT Biological systems which inspire connectionist architecture are based upon axonal point to point connections. As we attempt to engineer ever larger analogues of these neural networks we are forced to multiplex neural signals over time shared paths. This can alter the timing of signal arrival which is critical in oscillatory networks and leads to the bandwidth-synchrony dilemma. This paper further describes the problem, especially for analog VLSI systems, and approaches under consideration for the solution
Journal of Consciousness Exploration & Research, 2010
For the reasons discussed herein, it makes sense to treat consciousness as a process pervasive in... more For the reasons discussed herein, it makes sense to treat consciousness as a process pervasive in nature, at all levels of complexity. It can be seen as having a type of self-similarity. Recall that time supervenes on change, change requires contrast, and the contrast has to be detected. Whatever systems are changing are sensing and recording their reaction to the contrast in their behavior and in their state change.
A multi-chip analog neural network system capable of prototyping networks with as many as 81920 s... more A multi-chip analog neural network system capable of prototyping networks with as many as 81920 synaptic connections and 1024 neurons is described. The neural network architecture is reconfigurable by routing all neuron activation values through a host computer which can re-map the network connectivity by changing a look-up table in memory. Once a network is successfully prototyped, it is hardwired
My Editorial for this issue functions as an abstract: "Time & Experience: Twins of the Eternal No... more My Editorial for this issue functions as an abstract: "Time & Experience: Twins of the Eternal Now?" is located amongst my papers:
Uploads
Papers by Steve Deiss
http://unbc.academia.edu/GregoryNixon/Papers/247835/_Time_and_Experience_Twins_of_the_Eternal_Now_