Toward the biological model of the hippocampus as the successor representation agent

@article{Lee2020TowardTB,
  title={Toward the biological model of the hippocampus as the successor representation agent},
  author={Hyunsu Lee},
  journal={Bio Systems},
  year={2020},
  pages={
          104612
        }
}
  • Hyunsu Lee
  • Published 22 June 2020
  • Biology, Computer Science
  • Bio Systems

Figures from this paper

Neural learning rules for generating flexible predictions and computing the successor representation

The dynamics of a recurrent neural network naturally calculate the successor representation when the synaptic weights match the transition probability matrix, suggesting that the SR is more accessible in neural circuits than previously thought and can support a broad range of cognitive functions.

The effect of synaptic weight initialization in feature-based successor representation learning

This paper exposed SF learners to simple maze environments to analyze SF learning efficiency and W patterns pattern changes, and discussed the neurobiological meaning of SF weight matrix.

Correcting the hebbian mistake: Toward a fully error-driven hippocampus

It is shown that using a different form of learning based on correcting errors (error-driven learning) results in significantly improved episodic memory function in a biologically-based computational model of the hippocampus.

References

SHOWING 1-10 OF 40 REFERENCES

The hippocampus as a predictive map

It is argued that entorhinal grid cells encode a low-dimensionality basis set for the predictive representation, useful for suppressing noise in predictions and extracting multiscale structure for hierarchical planning.

Multiscale representation of very large environments in the hippocampus of flying bats

The study studied bats flying in a 200-m-long tunnel while the activity of hippocampal dorsal CA1 neurons was recorded using a custom wireless-electrophysiology system, finding that place cells recorded in the large environment exhibited a multifield, multiscale representation of space.

Scalable representation of time in the hippocampus

CA1 neuronal activity during temporal bisection tasks, in which the sets of time intervals to be discriminated were designed to be extended or contracted across the blocks of trials, demonstrated that scalable features of time cells may support the capability of flexible temporal representation for memory formation.

Frontal cortex neuron types categorically encode single decision variables

It is proposed that neurons in frontal cortex, as in other cortical regions, form a sparse and overcomplete representation of features relevant to the region’s function, and that they distribute this information selectively to downstream regions to support behaviour.

Grandmaster level in StarCraft II using multi-agent reinforcement learning

The agent, AlphaStar, is evaluated, which uses a multi-agent reinforcement learning algorithm and has reached Grandmaster level, ranking among the top 0.2% of human players for the real-time strategy game StarCraft II.

A neurally plausible model learns successor representations in partially observable environments

It is shown that distributional successor features can support reinforcement learning in noisy environments in which direct learning of successful policies is infeasible and which allows for efficient value function computation in partially observed environments via the successor representation.

Precise excitation-inhibition balance controls gain and timing in the hippocampus

Using patterned optogenetic stimulation of hippocampal CA3 neurons, it is suggested that this combination of precise EI balance and dynamic EI delays forms a general mechanism for millisecond-range input gating and subthreshold gain control in feedforward networks.

Predictive reward signal of dopamine neurons.

  • W. Schultz
  • Biology, Psychology
    Journal of neurophysiology
  • 1998
Dopamine systems may have two functions, the phasic transmission of reward information and the tonic enabling of postsynaptic neurons.