Canonical neural networks perform active inference

This work considers a class of canonical neural networks comprising rate coding models, wherein neural activity and plasticity minimise a common cost function—and plasticity is modulated with a certain delay. We show that such neural networks implicitly perform active inference and learning to minimise the risk associated with future outcomes. Mathematical analyses demonstrate that this biological optimisation can be cast as maximisation of model evidence, or equivalently minimisation of variational free energy, under the well-known form of a partially observed Markov decision process model. This equivalence indicates that the delayed modulation of Hebbian plasticity—accompanied with adaptation of firing thresholds—is a sufficient neuronal substrate to attain Bayes optimal inference and control. We corroborated this proposition using numerical analyses of maze tasks. This theory offers a universal characterisation of canonical neural networks in terms of Bayesian belief updating and provides insight into the neuronal mechanisms underlying planning and adaptive behavioural control.

The Shape of Data: a Theory of the Representation of Information in the Cerebellar Cortex

This paper presents a model of rate coding in the cerebellar cortex. The pathway of input to output of the cerebellum forms an anatomically repeating, functionally modular network, whose basic wiring is preserved across vertebrate taxa. Each network is bisected centrally by a functionally defined cell group, a microzone, which forms part of the cerebellar circuit. Input to a network may be from tens of thousands of concurrently active mossy fibres. The model claims to quantify the conversion of input rates into the code received by a microzone. Recoding on entry converts input rates into an internal code which is homogenised in the functional equivalent of an imaginary plane, occupied by the centrally positioned microzone. Homogenised means the code exists in any random sample of parallel fibre signals over a minimum number. The nature of the code and the regimented architecture of the cerebellar cortex mean that the threshold can be represented by space so that the threshold can be met by the physical dimensions of the Purkinje cell dendritic arbour and planar interneuron networks. As a result, the whole population of a microzone receives the same code. This is part of a mechanism which orchestrates functionally indivisible behaviour of the cerebellar circuit and is necessary for coordinated control of the output cells of the circuit. In this model, fine control of Purkinje cells is by input rates to the system and not by learning so that it is in conflict with the for-years-dominant supervised learning model.

Integration of neural architecture within a finite element framework for improved neuromusculoskeletal modeling

Neuromusculoskeletal (NMS) models can aid in studying the impacts of the nervous and musculoskeletal systems on one another. These computational models facilitate studies investigating mechanisms and treatment of musculoskeletal and neurodegenerative conditions. In this study, we present a predictive NMS model that uses an embedded neural architecture within a finite element (FE) framework to simulate muscle activation. A previously developed neuromuscular model of a motor neuron was embedded into a simple FE musculoskeletal model. Input stimulation profiles from literature were simulated in the FE NMS model to verify effective integration of the software platforms. Motor unit recruitment and rate coding capabilities of the model were evaluated. The integrated model reproduced previously published output muscle forces with an average error of 0.0435 N. The integrated model effectively demonstrated motor unit recruitment and rate coding in the physiological range based upon motor unit discharge rates and muscle force output. The combined capability of a predictive NMS model within a FE framework can aid in improving our understanding of how the nervous and musculoskeletal systems work together. While this study focused on a simple FE application, the framework presented here easily accommodates increased complexity in the neuromuscular model, the FE simulation, or both.

Synchrony-Division Neural Multiplexing: An Encoding Model

Cortical neurons receive mixed information from collective spiking activities of primary sensory neurons in response to a sensory stimulus. A recent study demonstrated that the time underlying the onset-offset of a tactile stimulus and its varying intensity can be respectively represented by synchronous and asynchronous spikes of S1 neurons in rats. This evidence capitalized on the ability of an ensemble of homogeneous neurons to multiplex, a coding strategy that was referred to as synchrony division multiplexing (SDM). Although neural multiplexing can be conceived by distinct functions of individual neurons in a heterogeneous neural ensemble, the extent to which nearly identical neurons in a homogeneous neural ensemble encode multiple features of a mixed stimulus remains unknown. Here, we present a computational framework to provide a system-level understanding of how an ensemble of homogeneous neurons enables SDM. First, we simulate SDM with an ensemble of homogeneous conductance-based model neurons receiving a mixed stimulus comprising slow and fast features. Using feature estimation techniques, we show that both features of the stimulus can be inferred from the generated spikes. Second, we utilize linear nonlinear (LNL) cascade models and calculate temporal filters and static nonlinearities of differentially synchronized spikes. We demonstrate that these filters and nonlinearities are distinct for synchronous and asynchronous spikes. Finally, we develop an augmented LNL cascade model as an encoding model for the SDM by combining individual LNLs calculated for each type of spike. The augmented LNL model reveals that a homogeneous neural ensemble can perform two different functions, namely, temporal- and rate- coding, simultaneously.

Across-Area Synchronization Supports Feature Integration in a Biophysical Network Model of Working Memory

Working memory function is severely limited. One key limitation that constrains the ability to maintain multiple items in working memory simultaneously is so-called swap errors. These errors occur when an inaccurate response is in fact accurate relative to a non-target stimulus, reflecting the failure to maintain the appropriate association or “binding” between the features that define one object (e.g., color and location). The mechanisms underlying feature binding in working memory remain unknown. Here, we tested the hypothesis that features are bound in memory through synchrony across feature-specific neural assemblies. We built a biophysical neural network model composed of two one-dimensional attractor networks – one for color and one for location – simulating feature storage in different cortical areas. Within each area, gamma oscillations were induced during bump attractor activity through the interplay of fast recurrent excitation and slower feedback inhibition. As a result, different memorized items were held at different phases of the network’s oscillation. These two areas were then reciprocally connected via weak cortico-cortical excitation, accomplishing binding between color and location through the synchronization of pairs of bumps across the two areas. Encoding and decoding of color-location associations was accomplished through rate coding, overcoming a long-standing limitation of binding through synchrony. In some simulations, swap errors arose: “color bumps” abruptly changed their phase relationship with “location bumps.” This model, which leverages the explanatory power of similar attractor models, specifies a plausible mechanism for feature binding and makes specific predictions about swap errors that are testable at behavioral and neurophysiological levels.

Physiological noise optimizes multiplexed coding of vibrotactile-like signals in somatosensory cortex

ABSTRACTNeurons can use different aspects of their spiking to simultaneously represent (multiplex) different features of a stimulus. For example, some pyramidal neurons in primary somatosensory cortex (S1) use the rate and timing of their spikes to respectively encode the intensity and frequency of vibrotactile stimuli. Doing so has several requirements. Because they fire at low rates, pyramidal neurons cannot entrain 1:1 with high-frequency (100-600 Hz) inputs and instead must skip (i.e. not respond to) some stimulus cycles. The proportion of skipped cycles must vary inversely with stimulus intensity for firing rate to encode stimulus intensity. Spikes must phase lock to the stimulus for spike times (intervals) to encode stimulus frequency but, in addition, skipping must occur irregularly to avoid aliasing. Using simulations and in vitro experiments in which S1 pyramidal neurons were stimulated with inputs emulating those induced by vibrotactile stimuli, we show that fewer cycles are skipped as stimulus intensity increases, as required for rate coding, and that physiological noise induces irregular skipping without disrupting phase locking, as required for temporal coding. This occurs because the reliability and precision of spikes evoked by small- amplitude, fast-onset signals are differentially sensitive to noise. Simulations confirmed that differences in stimulus intensity and frequency can be well discriminated based on differences in spike rate or timing, respectively, but only in the presence of noise. Our results show that multiplexed coding by S1 pyramidal neurons is facilitated rather than degraded by physiological levels of noise. In fact, multiplexing is optimal under physiologically noisy conditions.

On Error Probability of CRC/Polar Codes with List Decoding

<p>In this work the study has been done on the performance pertaining to the binary antipodal AWGN channel for rate ½ coding approaches, with short and medium blocklengths. This work emphasizes on the contributions of various events leading to block error and their dependence on signal-to-noise ratio, decoder list size, CRC length, if any, and the role of list sorting. Furthermore, this work generalizes to variations, including Reed-Muller/Polar codes that follow the polarization and the method of successive decoding.<br></p>

On Error Probability of CRC/Polar Codes with List Decoding

<p>In this work the study has been done on the performance pertaining to the binary antipodal AWGN channel for rate ½ coding approaches, with short and medium blocklengths. This work emphasizes on the contributions of various events leading to block error and their dependence on signal-to-noise ratio, decoder list size, CRC length, if any, and the role of list sorting. Furthermore, this work generalizes to variations, including Reed-Muller/Polar codes that follow the polarization and the method of successive decoding.<br></p>

Parallel pathways for rapid odor processing in lateral entorhinal cortex: Rate and temporal coding by layer 2 subcircuits

SummaryOlfactory information is encoded in lateral entorhinal cortex (LEC) by two classes of layer 2 (L2) principal neurons: fan and pyramidal cells. However, the functional properties of L2 neurons are unclear. Here, we show in awake mice that L2 cells respond rapidly to odors during single sniffs and that LEC is essential for discrimination of odor identity and intensity. Population analyses of L2 ensembles reveals that while rate coding distinguishes odor identity, firing rates are weakly concentration-dependent and changes in spike timing represent odor intensity. L2 principal cells differ in afferent olfactory input and connectivity with local inhibitory circuits and the relative timing of pyramidal and fan cell spikes underlies odor intensity coding. Downstream, intensity is encoded purely by spike timing in hippocampal CA1. Together, these results reveal the unique processing of odor information by parallel LEC subcircuits and highlight the importance of temporal coding in higher olfactory areas.

Across-area synchronization supports feature integration in working memory

Working memory function is severely limited. One key limitation that constrains the ability to maintain multiple items in working memory simultaneously is so-called swap errors. These errors occur when an inaccurate response is in fact accurate relative to a non-target stimulus, reflecting the failure to maintain the appropriate association or 'binding' between the features that define one object (e.g., color and location). The mechanisms underlying feature binding in working memory remain unknown. Here, we tested the hypothesis that features are bound in memory through synchrony across feature-specific neural assemblies. We built a biophysical neural network model composed of two one-dimensional attractor networks - one for color and one for location - simulating feature storage in different cortical areas. Within each area, gamma oscillations were induced during bump attractor activity through the interplay of fast recurrent excitation and slower feedback inhibition. As a result, different memorized items were held at different phases of the network's oscillation. These two areas were then reciprocally connected via weak cortico-cortical excitation, accomplishing binding between color and location through the synchronization of pairs of bumps across the two areas. Encoding and decoding of color-location associations was accomplished through rate coding, overcoming a long-standing limitation of binding through synchrony. In some simulations, swap errors arose: 'color bumps' abruptly changed their phase relationship with 'location bumps'. This model, which leverages the explanatory power of similar attractor models, specifies a plausible mechanism for feature binding and makes specific predictions about swap errors that are testable at behavioral and neurophysiological levels.