Continual Learning in a Multi-Layer Network of an Electric Fish

Muller, Salomon; Zadina, Abigail N.; Abbott, L. F.; Sawtell, Nathaniel B.

doi:10.1016/j.cell.2019.10.020

Cited by 26 publications

(7 citation statements)

References 55 publications

(78 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the cerebellum-like circuit of the electric fish, MG-cells, which are analogous to P-cells, produce broad spikes that are analogous to complex spikes. An unexpected sensory input produces broad spikes in some MGcells, but suppresses the broad spikes in others (Muller et al, 2019). Thus, MG-cells in the electric fish also exhibit a tuning for error.…”

Section: Population Coding Of P-cellsmentioning

confidence: 99%

Population coding in the cerebellum: a machine learning perspective

Shadmehr¹

2020

Journal of Neurophysiology

View full text Add to dashboard Cite

The cerebellum resembles a feedforward, three-layer network of neurons in which the "hidden layer" consists of Purkinje cells (P-cells), and the output layer consists of deep cerebellar nucleus (DCN) neurons. In this analogy, the output of each DCN neuron is a prediction that is compared to the actual observation, resulting in an error signal that originates in the inferior olive. Efficient learning requires that the error signal reach the DCN neurons, as well as the P-cells that project onto them. However, this basic rule of learning is violated in the cerebellum: the olivary projections to the DCN are weak, particularly in adulthood. Instead, an extraordinarily strong signal is sent from the olive to the P-cells, producing complex spikes. Curiously, P-cells are grouped into small populations that converge onto single DCN neurons. Why are the P-cells organized in this way, and what is the membership criterion of each population? Here, I apply elementary mathematics from machine learning and consider the fact that P-cells that form a population exhibit a special property: they can synchronize their complex spikes, which in turn suppress activity of DCN neuron they project to. Thus, complex spikes can not only act as a teaching signal for a P-cell, but through complex spike synchrony a P-cell population may act as a surrogate teacher for the DCN neuron that produced the erroneous output. It appears that grouping of P-cells into small populations that share a preference for error satisfies a critical requirement of efficient learning: providing error information to the output layer neuron (DCN) that was responsible for the error, as well as the hidden layer neurons (P-cells) that contributed to it. This population coding may account for several remarkable features of behavior during learning, including multiple timescales, protection from erasure, and spontaneous recovery of memory.

show abstract

Section: Population Coding Of P-cellsmentioning

confidence: 99%

Population coding in the cerebellum: a machine learning perspective

Shadmehr¹

2020

Journal of Neurophysiology

View full text Add to dashboard Cite

show abstract

“…In these electrolocation systems, corollary discharge functions to subtract predictive sensory feedback from reafferent input by a modifiable efference copy, forming a "negative image" through cerebellum-like circuitry in the electrosensory lateral line lobe (24,55,56). This negative image is formed at synapses between parallel fibers that carry corollary discharge inputs from the CN and medial ganglion cells that receive sensory inputs through spike-timing-dependent plasticity (59)(60)(61)(62). By contrast, in the communication pathway, we found that removal of sensory feedback had no effect on hormonal modulations of corollary discharge (Fig.…”

Section: Discussionmentioning

confidence: 99%

Hormonal coordination of peripheral motor output and corollary discharge in a communication system

Fukutomi

Carlson

2023

Preprint

View full text Add to dashboard Cite

Steroid hormones remodel neural networks to induce developmental or seasonal changes in animal behavior, but little is known about hormonal modulation of sensorimotor integration. Here, we investigate hormonal effects on a predictive motor signal, termed corollary discharge, that modulates sensory processing in weakly electric mormyrid fish. In the electrosensory pathway mediating communication behavior, inhibition activated by a corollary discharge precisely blocks sensory responses to self-generated electric pulses, allowing the downstream circuit to selectively analyze communication signals from nearby fish. These electric pulses are elongated by increasing testosterone levels in males during the breeding season. Using systematic testosterone treatment, we induced electric-pulse elongation in fish and found that the timing of electroreceptor spiking responses to self-generated pulses (reafference) was delayed as electric pulse duration increased. Recording evoked potentials from a midbrain electrosensory nucleus revealed that the timing of corollary discharge inhibition was delayed and elongated by testosterone. Further, this shift in corollary discharge timing was precisely matched to the shift in timing of the reafferent spikes. We then asked whether the shift in inhibition timing was caused by direct action of testosterone on the corollary discharge circuit or plasticity of the circuit through altered sensory feedback. We surgically silenced the electric organs of fish and found similar hormonal modulation of corollary discharge timing between intact and silent fish, suggesting that sensory feedback was not required for this shift. These results demonstrate that testosterone directly and independently modulates peripheral motor output and a predictive motor signal in a coordinated manner.

show abstract

“…The targets were smoothed broad-spike responses of 15 MG cells time-locked to an electric organ discharge command measured during experiments ( Muller et al, 2019 ). The original data set consisted of 55 MG cells, each with a 300ms long spike raster with a 1ms time bin.…”

Section: Methodsmentioning

confidence: 99%

Task-dependent optimal representations for cerebellar learning

Xie,

Muscinelli,

Decker Harris

et al. 2023

eLife

View full text Add to dashboard Cite

The cerebellar granule cell layer has inspired numerous theoretical models of neural representations that support learned behaviors, beginning with the work of Marr and Albus. In these models, granule cells form a sparse, combinatorial encoding of diverse sensorimotor inputs. Such sparse representations are optimal for learning to discriminate random stimuli. However, recent observations of dense, low-dimensional activity across granule cells have called into question the role of sparse coding in these neurons. Here, we generalize theories of cerebellar learning to determine the optimal granule cell representation for tasks beyond random stimulus discrimination, including continuous input-output transformations as required for smooth motor control. We show that for such tasks, the optimal granule cell representation is substantially denser than predicted by classical theories. Our results provide a general theory of learning in cerebellum-like systems and suggest that optimal cerebellar representations are task-dependent.

show abstract

Continual Learning in a Multi-Layer Network of an Electric Fish

Abstract: Highlights d Biological solutions to general problems in multi-layer learning are shown d Intermediate layer function requires compartmentalization of learning and signaling d Circuit organization based on learning solves the credit assignment problem

Cited by 26 publications

References 55 publications

Population coding in the cerebellum: a machine learning perspective

Population coding in the cerebellum: a machine learning perspective

Hormonal coordination of peripheral motor output and corollary discharge in a communication system

Task-dependent optimal representations for cerebellar learning

Contact Info

Product

Resources

About