Terrence C. Stewart

• University of Waterloo

Neuromorphic computing shows promise for advancing computing efficiency and capabilities of AI applications using brain-inspired principles. However, the neuromorphic research field currently lacks standardized benchmarks, making it difficult to accurately measure technological advancements, compare performance with conventional methods, and identi...

Spiking neural networks and neuromorphic hardware platforms that simulate neuronal dynamics are getting wide attention and are being applied to many relevant problems using Machine Learning. Despite a well-established mathematical foundation for neural dynamics, there exists numerous software and hardware solutions and stacks whose variability make...

Time is a central dimension against which perception, action, and cognition play out. From anticipating when future events will happen to recalling how long ago previous events occurred, humans and animals are exquisitely sensitive to temporal structure. Empirical evidence seems to suggest that estimating time prospectively (i.e., in passing) is qu...

The field of neuromorphic computing holds great promise in terms of advancing computing efficiency and capabilities by following brain-inspired principles. However, the rich diversity of techniques employed in neuromorphic research has resulted in a lack of clear standards for benchmarking, hindering effective evaluation of the advantages and stren...

This study proposes voltage-dependent-synaptic plasticity (VDSP), a novel brain-inspired unsupervised local learning rule for the online implementation of Hebb’s plasticity mechanism on neuromorphic hardware. The proposed VDSP learning rule updates the synaptic conductance on the spike of the postsynaptic neuron only, which reduces by a factor of t...

The International Conference on Cognitive Modeling brings together researchers from around the world whose main goal is to build computational systems that reflect the internal processes of the mind. In this issue, we present the five best representative papers on this work from our 19th meeting, ICCM 2021, which was held virtually from July 3 to J...

This study proposes voltage-dependent-synaptic plasticity (VDSP), a novel brain-inspired unsupervised local learning rule for the online implementation of Hebb's plasticity mechanism on neuromorphic hardware. The proposed VDSP learning rule updates the synaptic conductance on the spike of the postsynaptic neuron only, which reduces by a factor of t...

Mutual information (MI) is a standard objective function for driving exploration. The use of Gaussian processes to compute information gain is limited by time and memory complexity that grows with the number of observations collected.We present an efficient implementation of MI-driven exploration by combining vector symbolic architectures with Baye...

While neural networks are highly effective at learning task-relevant representations from data, they typically do not learn representations with the kind of symbolic structure that is hypothesized to support high-level cognitive processes, nor do they naturally model such structures within problem domains that are continuous in space and time. To f...

We implemented two neural network based benchmark tasks on a prototype chip of the second-generation SpiNNaker (SpiNNaker 2) neuromorphic system: keyword spotting and adaptive robotic control. Keyword spotting is commonly used in smart speakers to listen for wake words, and adaptive control is used in robotic applications to adapt to unknown dynami...

The International Conference on Cognitive Modeling brings together researchers from around the world whose main goal is to build computational systems that reflect the internal processes of the mind. In this issue, we present the four best representative papers on this work from our 18th meeting, ICCM 2020, which was also the first meeting to be he...

Neurophysiology and neuroanatomy constrain the set of possible computations that can be performed in a brain circuit. While detailed data on brain microcircuits is sometimes available, cognitive modelers are seldom in a position to take these constraints into account. One reason for this is the intrinsic complexity of accounting for biological mech...

Our understanding of the neurofunctional mechanisms of speech production and their pathologies is still incomplete. In this paper, a comprehensive model of speech production based on the Neural Engineering Framework (NEF) is presented. This model is able to activate sensorimotor plans based on cognitive-functional processes (i.e., generation of the...

Predicting future behavior of other traffic participants is an essential task that needs to be solved by automated vehicles and human drivers alike to achieve safe and situationaware driving. Modern approaches to vehicles trajectory prediction typically rely on data-driven models like neural networks, in particular LSTMs (Long Short-Term Memorys),...

Vector Symbolic Architectures belong to a family of related cognitive modeling approaches that encode symbols and structures in high-dimensional vectors. Similar to human subjects, whose capacity to process and store information or concepts in short-term memory is subject to numerical restrictions,the capacity of information that can be encoded in...

We implemented two neural network based benchmark tasks on a prototype chip of the second-generation SpiNNaker (SpiNNaker 2) neuromorphic system: keyword spotting and adaptive robotic control. Keyword spotting is commonly used in smart speakers to listen for wake words, and adaptive control is used in robotic applications to adapt to unknown dynami...

The interaction between robots and humans is of great relevance for the field of neurorobotics as it can provide insights on how humans perform motor control and sensor processing and on how it can be applied to robotics. We propose a spiking neural network (SNN) to trigger finger motion reflexes on a robotic hand based on human sEMG data. The firs...

Vector Symbolic Architectures belong to a family of related cognitive modeling approaches that encode symbols and structures in high-dimensional vectors. Similar to human subjects, whose capacity to process and store information or concepts in short-term memory is subject to numerical restrictions, the capacity of information that can be encoded in...

In this paper, we present an anomaly detection system employing an unsupervised learning model trained on the information encapsulated within distributed vector representations of automotive scenes. Our representations allows us to encode automotive scenes with a varying number of traffic participants in a vector of fixed length. We train a neural...

Predicting future behavior of other traffic participants is an essential task that needs to be solved by automated vehicles and human drivers alike to achieve safe and situation-aware driving. Modern approaches to vehicles trajectory prediction typically rely on data-driven models like neural networks, in particular LSTMs (Long Short-Term Memorys),...

Predicting future behavior and positions of other traffic participants from observations is a key problem that needs to be solved by human drivers and automated vehicles alike to safely navigate their environment and to reach their desired goal. In this paper, we expand on previous work on an automotive environment model based on vector symbolic ar...

Most computational models of timing rely on well-defined start- and stop-signals, however, these are quite rare in our natural environments. Moreover, theories typically propose different mechanisms to account for retrospective and prospective timing, an assumption that is difficult to align with naturalistic, continuative types of timing. Here we...

Predicting future motion of other vehicles or, more generally, the development of traffic situations, is an essential step towards secure, context-aware automated driving. On the one hand, human drivers are able to anticipate driving situations continuously based on the currently perceived behavior of other traffic participants while incorporating...

We present a novel method for constructing neurally implemented spatial representations that we show to be useful for building models of spatial cognition. This method represents continuous (i.e., real-valued) spaces using neurons, and identifies a set of operations for manipulating these representations. Specifically, we use "fractional binding" t...

We propose a spiking recurrent neural network model of flexible human timing behavior based on the delay network. The well-known 'scalar property' of timing behavior arises from the model in a natural way, and critically depends on how many dimensions are used to represent the history of stimuli. The model also produces heterogeneous firing pattern...

Emotion theory needs to explain the relationship of language and emotions, and the embodiment of emotions, by specifying the computational mechanisms underlying emotion generation in the brain. We used Chris Eliasmith's Semantic Pointer Architecture to develop POEM, a computational model that explains numerous important phenomena concerning emotion...

Predicting future vehicle behaviour is an essential task to enable safe and situation-aware automated driving. In this paper, we propose to encapsulate spatial information of multiple objects in a semantic vector-representation. Assuming that future vehicle motion is influenced not only by past positions but also by the behaviour of other traffic p...

Braindrop is the first neuromorphic system designed to be programmed at a high level of abstraction. Previous neuromorphic systems were programmed at the neurosynaptic level and required expert knowledge of the hardware to use. In stark contrast, Braindrop’s computations are specified as coupled nonlinear dynamical systems and synthesized to the ha...

Background: Parkinson's disease affects many motor processes including speech. Besides drug treatment, deep brain stimulation (DBS) in the subthalamic nucleus (STN) and globus pallidus internus (GPi) has developed as an effective therapy. Goal: We present a neural model that simulates a syllable repetition task and evaluate its performance when var...

In this paper, we propose a novel approach to knowledge representation for automotive environment modelling based on Vector Symbolic Architectures (VSAs). We build a vector representation describing structured information and relations within the current scene based on high-level object-lists perceived by individual sensors. Such a representation c...

We propose a neuromorphic approach to perception, reasoning and motor control using Spiking Neural Networks in mobile robotics. We demonstrate this by using a mobile robotic manipulator solving a pick-and-place task. All sensory data is provided by spike-based silicon retina cameras - eDVS (embedded Dynamic Vision Sensor) - and all reasoning and mo...

We present a novel approach to achieving temperature-robust behavior in neuromorphic systems that operates at the population level, trading an increase in silicon-neuron count for robustness across temperature. Our silicon neurons' tuning curves were highly sensitive to temperature, which could be decoded from a 400-neuron population with a precisi...

Generating associations is important for cognitive tasks including language acquisition and creative problem solving. It remains an open question how the brain represents and processes associations. The Remote Associates Test (RAT) is a task, originally used in creativity research, that is heavily dependent on generating associations in a search fo...

We provide a short proof that the uniform distribution of points for the n-ball is equivalent to the uniform distribution of points for the (n + 1)-sphere projected onto n dimensions. This implies the surprising result that one may uniformly sample the n-ball by instead uniformly sampling the (n + 1)-sphere and then arbitrarily discarding two coord...

We use a spiking neural network model of working memory (WM) capable of performing the spatial delayed response task (DRT) to investigate two drugs that affect WM: guanfacine (GFC) and phenylephrine (PHE). In this model, the loss of information over time results from changes in the spiking neural activity through recurrent connections. We reproduce...

We present a spiking neuron model of the motor cortices and cerebellum of the motor control system. The model consists of anatomically organized spiking neurons encompassing premotor, primary motor, and cerebellar cortices. The model proposes novel neural computations within these areas to control a nonlinear three-link arm model that can adapt to...

The ability to associate words is an important cognitive skill. In this study we investigate different methods for representing word associations in the brain, using the Remote Associates Test (RAT) as a task. We explore representations derived from free association norms and statistical n-gram data. Although n-gram representations yield better per...

In this paper, we present a spiking neural model of context dependent decision making. Prefrontal cortex (PFC) plays a fundamental role in context dependent behaviour. We model the PFC at the level of single spiking neurons, to explore the underlying computations which determine its contextual responses. The model is built using the Neural Engineer...

We demonstrate a hybrid neuromorphic learning paradigm that learns complex sensorimotor mappings based on a small set of hard-coded reflex behaviours. A mobile robot is first controlled by a basic set of reflexive hand-designed behaviours. All sensor data is provided via a spike-based silicon retina camera (eDVS), and all control is implemented via...

Further details of the paper, including: comparison to past work, predictions made, methods and materials, model architecture, and neural analysis.

Further details of the paper, including: comparison to past work, predictions made, methods and materials, model architecture, and neural analysis.

Further details of the paper, including: comparison to past work, predictions made, methods and materials, model architecture, and neural analysis.

We present a spiking neuron model of the motor cortices and cerebellum of the motor control system. The model consists of anatomically organized spiking neurons encompassing premotor, primary motor and cerebellar cortices. The model proposes novel neural computations within these areas to control a nonlinear three-link arm model that can adapt to u...

Evaluating the effectiveness and performance of neuromorphic hardware is difficult. It is even more difficult when the task of interest is a closed-loop task; that is, a task where the output from the neuromorphic hardware affects some environment, which then in turn affects the hardware's future input. However, closed-loop situations are one of th...

Nengo is a software package for designing and simulating large-scale neural models. Nengo is architected such that the same Nengo model can be simulated on any of several Nengo backends with few to no modifications. Backends translate a model to specific platforms, which include GPUs and neuromorphic hardware. Nengo also contains a large test suite...

Living organisms are capable of autonomously adapting to dynamically changing environments by receiving inputs from highly specialized sensory organs and elaborating them on the same parallel, power-efficient neural substrate. In this paper we present a prototype for a comprehensive integrated platform that allows replicating principles of neural i...

Visuospatial attention produces myriad effects on the activity and selectivity of cortical neurons. Spiking neuron models capable of reproducing a wide variety of these effects remain elusive. We present a model called the Attentional Routing Circuit (ARC) that provides a mechanistic description of selective attentional processing in cortex. The mo...

Neuroscience currently lacks a comprehensive theory of how cognitive processes can be implemented in a biological substrate. The Neural Engineering Framework (NEF) proposes one such theory, but has not yet gathered significant empirical support, partly due to the technical challenge of building and simulating large-scale models with the NEF. Nengo...

We propose a unified theory of intentions as neural processes that integrate representations of states of affairs, actions, and emotional evaluation. We show how this theory provides answers to philosophical questions about the concept of intention, psychological questions about human behavior, computational questions about the relations between be...

Quantum probability (QP) theory can be seen as a type of vector symbolic architecture (VSA): mental states are vectors storing structured information and manipulated using algebraic operations. Furthermore, the operations needed by QP match those in other VSAs. This allows existing biologically realistic neural models to be adapted to provide a mec...

We present a neural mechanism for interpreting and executing visually presented commands. These are simple verb-noun commands (such as WRITE THREE) and can also include conditionals ([if] SEE SEVEN, [then] WRITE THREE). We apply this to a simplified version of our large-scale functional brain model "Spaun", where input is a 28x28 pixel visual stimu...

Modeling the Brain Neurons are pretty complicated cells. They display an endless variety of shapes that sprout highly variable numbers of axons and dendrites; they sport time- and voltage-dependent ion channels along with an impressive array of neurotransmitter receptors; and they connect intimately with near neighbors as well as former neighbors w...

We will demonstrate a model of rat hippocampus place, grid and Border cells implemented with the SpiNNaker spiking neural hardware, configured using the Neural Engineering Framework (NEF) package Nengo, hosted on a mobile robot. These cells are used by rats for odometry, to locate landmarks and for navigation. Nengo provides a dynamical systems app...

We use neuromorphic chips to perform arbitrary mathematical computations for the first time. Static and dynamic computations are realized with heterogeneous spiking silicon neurons by programming their weighted connections. Using 4K neurons with 16M feed-forward or recurrent synaptic connections, formed by 256K local arbors, we communicate a scalar...

We expand our existing spiking neuron model of decision making in the cortex and basal ganglia to include local learning on the synaptic connections between the cortex and striatum, modulated by a dopaminergic reward signal. We then compare this model to animal data in the bandit task, which is used to test rodent learning in conditions involving f...

Methods for cleaning up (or recognizing) states of a neural network are crucial for the functioning of many neural cognitive models. For example, Vector Symbolic Architectures provide a method for manipulating symbols using a fixed-length vector representation. To recognize the result of these manipulations, a method for cleaning up the resulting n...

This paper re-examines the question of localist vs. distributed neural representations using a biologically realistic framework based on the central notion of neurons having a preferred direction vector.Apreferred direction vector captures the general observation that neurons fire most vigorously when the stimulus lies in a particular direction in...

We present a model of attentional routing called the Attentional Routing Circuit (ARC) that extends an existing model of spiking neurons with dendritic nonlinearities. Specifically, we employ the Poirazi et al. (2003) pyramidal neuron in a population coding framework. ARC demonstrates that the dendritic nonlinearities can be exploited to result in...

We present a computational model capable of solving arbitrary Tower of Hanoi problems. All elements except visual input and motor output are implemented using 150,000 LIF spiking neurons. Properties of these neurons (firing rate, post-synaptic time constant, etc.) are set based on the neurons in corresponding areas of the brain, and connectivity is...

Motor prostheses aim to restore function to disabled patients. Despite compelling proof of concept systems, barriers to clinical translation remain. One challenge is to develop a low-power, fully-implantable system that dissipates only minimal power so as not to damage tissue. To this end, we implemented a Kalman-filter based decoder via a spiking...

Many kinds of creativity result from combination of mental representations. This paper provides a computational account of how creative thinking can arise from combining neural patterns into ones that are potentially novel and useful. We defend the hypothesis that such combinations arise from mechanisms that bind together neural activity by a proce...

Testing for Equivalence: A Methodology for Computational Cognitive Modelling The equivalence test (Stewart and West, 2007; Stewart, 2007) is a statistical measure for evaluating the similarity between a model and the system being modelled. It is designed to avoid over-fitting and to generate an easily interpretable summary of the quality of a model...

We present a model of sym ol manipulation implemented usin! spi"in! neurons and closely tied to the anatomy of the corte#, asal !an!lia, and thalamus$ The model is a !eneral% purpose neural controller &hich plays a role analo!ous to a production system$ 'nformation stored in corte# is used y the asal !an!lia as the asis for selectin! et&een a set o...

A fundamental process for cognition is action selection: choosing a particular action out of the many possible actions available. This process is widely believed to involve the basal ganglia, and we present here a model of action selection that uses spiking neurons and is in accordance with the connectivity and neuron types found in this area. Sinc...

We have developed a computational model using spiking neurons that provides the decision-making capabilities required for production system models of cognition. This model conforms to the anatomy and connectivity of the basal ganglia, and the neuron parameters are set based on known neurophysiology. Behavioral-level timing and neural-level spike pr...

Erev, Ert, and Roth organized three choice prediction competitions focused on three related choice tasks: One shot decisions from description (decisions under risk), one shot decisions from experience, and repeated decisions from experience. Each competition was based on two experimental datasets: An estimation dataset, and a competition dataset. T...

Nengo (http://nengo.ca) is an open-source neural simulator that has been greatly enhanced by the recent addition of a Python script interface. Nengo provides a wide range of features that are useful for physiological simulations, including unique features that facilitate development of population-coding models using the neural engineering framework...

For the Technion Prediction Tournament, we developed a model of making repeated binary choices between a safe option and a risky option. The model is based on the ACT-R declarative memory system, with the use of the Blending mechanism and sequential dependencies. By using established cognitive theory, rather than specialized machine learning techni...

A common feature of many cognitive architectures is a central executive control with a 50-millisecond cycle time. This system determines which action to perform next, based on the current context. We present the first model of this system using spiking neurons. Given the constraints of well-established neural time constants, a cycle time of 46.6 mi...

Evaluating variations in the structure of computational models of cognition is as important as evaluating variations in the numerical parameters of such models. However, computational models tend not to be organized in such a way as to directly support such research. To address this need, we have taken the well-known cognitive architecture ACT-R, r...

Organisms across species use the strategy of generating structures in their environment to lower cog-nitive complexity. Examples include pheromones, markers, color codes, etc. We provide a model of how such structures originate, and present a simulation where organisms with only reactive behavior learn, within their lifetime, to add such structures...

A series of computational models are presented which address the question of how peer relations change over time. We examine data from a standardized metric (CDC) that places school children in one of five categories: Popular, Rejected, Neglected, Controversial, and Average, and how such classifications change over time. A simple random model is sh...

We provide two descriptions of the Q-Learning algorithm (Watkins, 1989), one high-level and the other at the mechanism-level, to support the use of the algorithm within cognitive modeling frameworks. High-level Description The Q-Learning algorithm is a probabilistic learning rule that maps states in the world [s] to possible actions [a]. For an upc...

Abstract Most courses ,on computational, modeling ,involve ,having students skilled in the ,computer ,science implement ,various computational models. Here, a course is presented which is aimed,at students ,with minimal ,programming ,background. The focus is on using a large number ,of different models,on various,problems. , Students ,learn ,how ,w...

There are many computational models whose broad purpose is to allow an agent to learn via experience to perform effectively in a given environment. However, it is uncommon to see these models directly compared to each other, or to empirical data of real creatures adapting to their environments. Here, a comparison methodology is proposed involving v...

A computational model of the formation of peer groups in children is presented. We used standard sociometric measurements (the CDC classification of Popular, Rejected, Neglected, and Controversial individuals) to compare the model to empirical data. The model fit this data well in terms of category distributions and stability, even without introduc...

We present a novel cognitive architecture built from realistic spiking neurons that exhibits basic production system capabilities. It uses Holographic Reduced Representations to encode structured information and the Neural Engineering Framework to create detailed neuro-biologically plausible networks of spiking neurons for storing and manipulating...

Tutorial Objectives As we learn more about the neural activity underlying cognitive function, there is an increasing demand to explicitly and quantitatively connect cognitive theories to neurological details. Bridging these levels provides benefits in both directions; aspects of the cognitive theory can predict and be constrained by neurological de...

A methodology is described for developing cognitive science theories which produce numerical predictions. This is done by adopting methodology from mathematical models in physics, and adapting it for use with the more complex computational models. Bootstrap confidence intervals and equivalence testing are introduced, and parameter fitting is shown...

Cognitive redeployment is the idea that an important part of the evolution of cognition is the adaptation and re-use of existing cognitive modules for new purposes. In this paper, we apply this idea to the ACT-R declarative memory system and the ACT-R visual system. We have developed a common underlying implementation of these systems as part of th...

As cognitive models are developed that are meant to apply to a broad range of phenomena, it is necessary to evaluate how successfully they do so. This is commonly done by measures such as the Mean Squared Error. We propose and demonstrate an alternate approach based on a measure of statistical equivalence. Instead of using sample means, this method...