• Corpus ID: 238857279

Sign and Relevance learning

@article{Daryanavard2021SignAR,
  title={Sign and Relevance learning},
  author={Sama Daryanavard and Bernd Porr},
  journal={ArXiv},
  year={2021},
  volume={abs/2110.07292}
}
Standard models of biologically realistic, or inspired, reinforcement learning employ a global error signal which implies shallow networks. However, deep networks could offer a drastically superior performance by feeding the error signal backwards through such a network which in turn is not biologically realistic as it requires symmetric weights between top-down and bottom-up pathways. Instead, we present a network combining local learning with global modulation where neuromodulation controls… 

References

SHOWING 1-10 OF 50 REFERENCES
Attention-Gated Brain Propagation: How the brain can implement reward-based error backpropagation
TLDR
Attention-Gated Brain Propagation – a biologically plausible reinforcement learning scheme for deep networks with an arbitrary number of layers that achieves an accuracy that is equivalent to that of standard error-backpropagation, and better than state-of-the-art biologically inspired learning schemes.
Toward an Integration of Deep Learning and Neuroscience
TLDR
It is argued that a range of implementations of credit assignment through multiple layers of neurons are compatible with current knowledge of neural circuitry, and that the brain's specialized systems can be interpreted as enabling efficient optimization for specific problem classes.
Random synaptic feedback weights support error backpropagation for deep learning
TLDR
A surprisingly simple mechanism that assigns blame by multiplying errors by even random synaptic weights is presented, which can transmit teaching signals across multiple layers of neurons and performs as effectively as backpropagation on a variety of tasks.
Strongly Improved Stability and Faster Convergence of Temporal Sequence Learning by Using Input Correlations Only
TLDR
Elimination of the output from the learning rule removes the unwanted, destabilizing autocorrelation term, allowing us to use high learning rates, and mathematically show that the theoretical optimum of one-shot learning can be reached under ideal conditions with the new rule.
Temporal Sequence Learning, Prediction, and Control: A Review of Different Models and Their Relation to Biological Mechanisms
TLDR
This review compares methods for temporal sequence learning (TSL) across the disciplines machine-control, classical conditioning, neuronal models for TSL as well as spike-timing-dependent plasticity (STDP) and focuses on to what degree are reward-based and correlation-based learning related.
Closed-Loop Deep Learning: Generating Forward Models With Backpropagation
TLDR
This work shows in z-space specifically how error backpropagation can be achieved and in general how gradient-based approaches can be analyzed in such closed-loop scenarios.
Learning with Relevance: Using a Third Factor to Stabilize Hebbian Learning
TLDR
A third factor that switches on learning when the autocorrelation is minimal or zero and the cross-correlation is maximal is shown, and it is shown in a behavioral experiment that this three-factor learning clearly outperforms classical Hebbian learning.
Is backpropagation biologically plausible?
  • D. Stork
  • Computer Science
    International 1989 Joint Conference on Neural Networks
  • 1989
TLDR
The authors finds that in several posited implementations these design considerations imply that a finely structured neural connectivity is needed as well as a number of neurons and synapses beyond those inferred from the algorithmic network presentations of backpropagation.
A Physiologically Plausible Model of Action Selection and Oscillatory Activity in the Basal Ganglia
TLDR
A new spiking neuron model of the BG circuitry supports motor program selection and switching, which deteriorates under dopamine-depleted and dopamine-excessive conditions in a manner consistent with some pathologies associated with those dopamine states.
Dopamine Neurons Respond to Errors in the Prediction of Sensory Features of Expected Rewards
TLDR
It is shown that classic error-signaling dopamine neurons also respond to changes in value-neutral sensory features of an expected reward, suggesting that dopamine neurons have access to a wider variety of information than contemplated by the models currently used to interpret their activity.
...
1
2
3
4
5
...