A Neural Network of Adaptively Timed Reinforcement Learning and Hippocampal Dynamics

A neural model is described of how adaptively timed reinforcement learning occurs. The adaptive timing circuit is suggested to exist in the hippocampus, and to involve convergence of dentate granule cells on CA3 pyramidal cells, and NMDA receptors. This circuit forms part of a model neural system for the coordinated control of recognition learning, reinforcement learning, and motor learning, whose properties clarify how an animal can learn to acquire a delayed reward. Behavioral and neural data are summarized in support of each processing stage of the system. The relevant anatomical sites are in thalamus, neocortex, hippocampus, hypothalamus, amygdala, and cerebellum. Cerebellar influences on motor learning are distinguished from hippocampal influences on adaptive timing of reinforcement learning. The model simulates how damage to the hippocampal formation disrupts adaptive timing, eliminates attentional blocking, and causes symptoms of medial temporal amnesia. It suggests how normal acquisition of subcortical emotional conditioning can occur after cortical ablation, even though extinction of emotional conditioning is retarded by cortical ablation. The model simulates how increasing the duration of an unconditioned stimulus increases the amplitude of emotional conditioning, but does not change adaptive timing; and how an increase in the intensity of a conditioned stimulus "speeds up the clock", but an increase in the intensity of an unconditioned stimulus does not. Computer simulations of the model fit parametric conditioning data, including a Weber law property and an inverted U property. Both primary and secondary adaptively timed conditioning are simulated, as are data concerning conditioning using multiple interstimulus intervals (ISIs), gradually or abruptly changing ISis, partial reinforcement, and multiple stimuli that lead to time-averaging of responses. Neurobiologically testable predictions are made to facilitate further tests of the model.

Equilibria and Dynamics of a Neural Network Model for Opponent Muscle Control

One of the advantages of biological skeleto-motor systems is the opponent muscle design, which in principle makes it possible to achieve facile independent control of joint angle and joint stiffness. Prior analysis of equilibrium states of a biologically-based neural network for opponent muscle control, the FLETE model, revealed that such independent control requires specialized interneuronal circuitry to efficiently coordinate the opponent force generators. In this chapter, we refine the FLETE circuit variables specification and update the equilibrium analysis. We also incorporate additional neuronal circuitry that ensures efficient opponent force generation and velocity regulation during movement.

Foundational Theory for Understanding Policy Routing Dynamics

In this paper we introduce a theory of policy routing dynamics based on fundamental axioms of routing update mechanisms. We develop a dynamic policy routing model (DPR) that extends the static formalism of the stable paths problem (introduced by Griffin et al.) with discrete synchronous time. DPR captures the propagation of path changes in any dynamic network irrespective of its time-varying topology. We introduce several novel structures such as causation chains, dispute fences and policy digraphs that model different aspects of routing dynamics and provide insight into how these dynamics manifest in a network. We exercise the practicality of the theoretical foundation provided by DPR with two fundamental problems: routing dynamics minimization and policy conflict detection. The dynamics minimization problem utilizes policy digraphs, that capture the dependencies in routing policies irrespective of underlying topology dynamics, to solve a graph optimization problem. This optimization problem explicitly minimizes the number of routing update messages in a dynamic network by optimally changing the path preferences of a minimal subset of nodes. The conflict detection problem, on the other hand, utilizes a theoretical result of DPR where the root cause of a causation cycle (i.e., cycle of routing update messages) can be precisely inferred as either a transient route flap or a dispute wheel (i.e., policy conflict). Using this result we develop SafetyPulse, a token-based distributed algorithm to detect policy conflicts in a dynamic network. SafetyPulse is privacy preserving, computationally efficient, and provably correct.

Principles of Safe Policy Routing Dynamics

We introduce the Dynamic Policy Routing (DPR) model that captures the propagation of route updates under arbitrary changes in topology or path preferences. DPR introduces the notion of causation chains where the route flap at one node causes a flap at the next node along the chain. Using DPR, we model the Gao-Rexford (economic) guidelines that guarantee the safety (i.e., convergence) of policy routing. We establish three principles of safe policy routing dynamics. The non-interference principle provides insight into which ASes can directly induce route changes in one another. The single cycle principle and the multi-tiered cycle principle provide insight into how cycles of routing updates can manifest in any network. We develop INTERFERENCEBEAT, a distributed algorithm that propagates a small token along causation chains to check adherence to these principles. To enhance the diagnosis power of INTERFERENCEBEAT, we model four violations of the Gao-Rexford guidelines (e.g., transiting between peers) and characterize the resulting dynamics.

Temporal Dynamics of Decision-Making during Motion Perception in the Visual Cortex

How does the brain make decisions? Speed and accuracy of perceptual decisions covary with certainty in the input, and correlate with the rate of evidence accumulation in parietal and frontal cortical "decision neurons." A biophysically realistic model of interactions within and between Retina/LGN and cortical areas V1, MT, MST, and LIP, gated by basal ganglia, simulates dynamic properties of decision-making in response to ambiguous visual motion stimuli used by Newsome, Shadlen, and colleagues in their neurophysiological experiments. The model clarifies how brain circuits that solve the aperture problem interact with a recurrent competitive network with self-normalizing choice properties to carry out probablistic decisions in real time. Some scientists claim that perception and decision-making can be described using Bayesian inference or related general statistical ideas, that estimate the optimal interpretation of the stimulus given priors and likelihoods. However, such concepts do not propose the neocortical mechanisms that enable perception, and make decisions. The present model explains behavioral and neurophysiological decision-making data without an appeal to Bayesian concepts and, unlike other existing models of these data, generates perceptual representations and choice dynamics in response to the experimental visual stimuli. Quantitative model simulations include the time course of LIP neuronal dynamics, as well as behavioral accuracy and reaction time properties, during both correct and error trials at different levels of input ambiguity in both fixed duration and reaction time tasks. Model MT/MST interactions compute the global direction of random dot motion stimuli, while model LIP computes the stochastic perceptual decision that leads to a saccadic eye movement.

Spiking Dynamics during Perceptual Grouping in the Laminar Circuits of Visual Cortex

Grouping of collinear boundary contours is a fundamental process during visual perception. Illusory contour completion vividly illustrates how stable perceptual boundaries interpolate between pairs of contour inducers, but do not extrapolate from a single inducer. Neural models have simulated how perceptual grouping occurs in laminar visual cortical circuits. These models predicted the existence of grouping cells that obey a bipole property whereby grouping can occur inwardly between pairs or greater numbers of similarly oriented and co-axial inducers, but not outwardly from individual inducers. These models have not, however, incorporated spiking dynamics. Perceptual grouping is a challenge for spiking cells because its properties of collinear facilitation and analog sensitivity to inducer configurations occur despite irregularities in spike timing across all the interacting cells. Other models have demonstrated spiking dynamics in laminar neocortical circuits, but not how perceptual grouping occurs. The current model begins to unify these two modeling streams by implementing a laminar cortical network of spiking cells whose intracellular temporal dynamics interact with recurrent intercellular spiking interactions to quantitatively simulate data from neurophysiological experiments about perceptual grouping, the structure of non-classical visual receptive fields, and gamma oscillations.

ART-EMAP: A Neural Network Architecture for Learning and Prediction by Evidence Accumulation

This paper introduces ART-EMAP, a neural architecture that uses spatial and temporal evidence accumulation to extend the capabilities of fuzzy ARTMAP. ART-EMAP combines supervised and unsupervised learning and a medium-term memory process to accomplish stable pattern category recognition in a noisy input environment. The ART-EMAP system features (i) distributed pattern registration at a view category field; (ii) a decision criterion for mapping between view and object categories which can delay categorization of ambiguous objects and trigger an evidence accumulation process when faced with a low confidence prediction; (iii) a process that accumulates evidence at a medium-term memory (MTM) field; and (iv) an unsupervised learning algorithm to fine-tune performance after a limited initial period of supervised network training. ART-EMAP dynamics are illustrated with a benchmark simulation example. Applications include 3-D object recognition from a series of ambiguous 2-D views.

A Neural Network Architecture for Autonomous Learning, Recognition, and Prediction in a Nonstationary World

In a constantly changing world, humans are adapted to alternate routinely between attending to familiar objects and testing hypotheses about novel ones. We can rapidly learn to recognize and narne novel objects without unselectively disrupting our memories of familiar ones. We can notice fine details that differentiate nearly identical objects and generalize across broad classes of dissimilar objects. This chapter describes a class of self-organizing neural network architectures--called ARTMAP-- that are capable of fast, yet stable, on-line recognition learning, hypothesis testing, and naming in response to an arbitrary stream of input patterns (Carpenter, Grossberg, Markuzon, Reynolds, and Rosen, 1992; Carpenter, Grossberg, and Reynolds, 1991). The intrinsic stability of ARTMAP allows the system to learn incrementally for an unlimited period of time. System stability properties can be traced to the structure of its learned memories, which encode clusters of attended features into its recognition categories, rather than slow averages of category inputs. The level of detail in the learned attentional focus is determined moment-by-moment, depending on predictive success: an error due to over-generalization automatically focuses attention on additional input details enough of which are learned in a new recognition category so that the predictive error will not be repeated. An ARTMAP system creates an evolving map between a variable number of learned categories that compress one feature space (e.g., visual features) to learned categories of another feature space (e.g., auditory features). Input vectors can be either binary or analog. Computational properties of the networks enable them to perform significantly better in benchmark studies than alternative machine learning, genetic algorithm, or neural network models. Some of the critical problems that challenge and constrain any such autonomous learning system will next be illustrated. Design principles that work together to solve these problems are then outlined. These principles are realized in the ARTMAP architecture, which is specified as an algorithm. Finally, ARTMAP dynamics are illustrated by means of a series of benchmark simulations.

Cortical Dynamics of Visual Motion Perception: Short-Range and Long Range Apparent Motion

This article describes further evidence for a new neural network theory of biological motion perception that is called a Motion Boundary Contour System. This theory clarifies why parallel streams Vl-> V2 and Vl-> MT exist for static form and motion form processing among the areas Vl, V2, and MT of visual cortex. The Motion Boundary Contour System consists of several parallel copies, such that each copy is activated by a different range of receptive field sizes. Each copy is further subdivided into two hierarchically organized subsystems: a Motion Oriented Contrast Filter, or MOC Filter, for preprocessing moving images; and a Cooperative-Competitive Feedback Loop, or CC Loop, for generating emergent boundary segmentations of the filtered signals. The present article uses the MOC Filter to explain a variety of classical and recent data about short-range and long-range apparent motion percepts that have not yet been explained by alternative models. These data include split motion; reverse-contrast gamma motion; delta motion; visual inertia; group motion in response to a reverse-contrast Ternus display at short interstimulus intervals; speed-up of motion velocity as interfiash distance increases or flash duration decreases; dependence of the transition from element motion to group motion on stimulus duration and size; various classical dependencies between flash duration, spatial separation, interstimulus interval, and motion threshold known as Korte's Laws; and dependence of motion strength on stimulus orientation and spatial frequency. These results supplement earlier explanations by the model of apparent motion data that other models have not explained; a recent proposed solution of the global aperture problem, including explanations of motion capture and induced motion; an explanation of how parallel cortical systems for static form perception and motion form perception may develop, including a demonstration that these parallel systems are variations on a common cortical design; an explanation of why the geometries of static form and motion form differ, in particular why opposite orientations differ by 90°, whereas opposite directions differ by 180°, and why a cortical stream Vl -> V2 -> MT is needed; and a summary of how the main properties of other motion perception models can be assimilated into different parts of the Motion Boundary Contour System design.

Cortical Dynamics of Visual Motion Perception: Short-Range and Long-Range Apparent Motion

This article describes further evidence for a new neural network theory of biological motion perception. The theory clarifies why parallel streams Vl --> V2, Vl --> MT, and Vl --> V2 --> MT exist for static form and motion form processing among the areas Vl, V2, and MT of visual cortex. The theory suggests that the static form system (Static BCS) generates emergent boundary segmentations whose outputs are insensitive to direction-ofcontrast and insensitive to direction-of-motion, whereas the motion form system (Motion BCS) generates emergent boundary segmentations whose outputs are insensitive to directionof-contrast but sensitive to direction-of-motion. The theory is used to explain classical and recent data about short-range and long-range apparent motion percepts that have not yet been explained by alternative models. These data include beta motion; split motion; gamma motion and reverse-contrast gamma motion; delta motion; visual inertia; the transition from group motion to element motion in response to a Ternus display as the interstimulus interval (ISI) decreases; group motion in response to a reverse-contrast Ternus display even at short ISIs; speed-up of motion velocity as interflash distance increases or flash duration decreases; dependence of the transition from element motion to group motion on stimulus duration and size; various classical dependencies between flash duration, spatial separation, ISI, and motion threshold known as Korte's Laws; dependence of motion strength on stimulus orientation and spatial frequency; short-range and long-range form-color interactions; and binocular interactions of flashes to different eyes.