Voltage-dependent reversal potentials in spiking recurrent neural networks enhance energy efficiency and task performance
Rodrigues, M.; Gasco-Galvez, C.; Vinck, M.
Show abstract
Spiking recurrent neural networks (SRNNs) rival gated RNNs on various tasks, yet they still lack several hallmarks of biological neural networks. We introduce a biologically grounded SRNN that implements Dales law with voltage-dependent AMPA and GABA reversal potentials. These reversal potentials modulate synaptic gain as a function of the postsynaptic membrane potential, and we derive theoretically how they make each neurons effective dynamics and subthreshold resonance input-dependent. We trained SRNNs on the Spiking Heidelberg Digits dataset, and show that SRNN with reversal potentials cuts spike energy by up to 4x, while increasing task accuracy. This leads to high-performing Dalean SRNNs, substantially improving on Dalean networks without reversal potentials. Thus, Dales law with reversal potentials, a core feature of biological neural networks, can render SRNNs more accurate and energy-efficient.
Matching journals
The top 4 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Emergence of Sparse Coding, Balance and Decorrelation from a Biologically-Grounded Spiking Neural Network Model of Learning in the Primary Visual Cortex 98%
- Unsupervised Learning of Perceptual Feature Combinations 98%
- Brain signal predictions from multi-scale networks using a linearized framework 97%
Similar papers in this journal
- Heterogeneity in Neuronal Dynamics is Learned by Gradient Descent for Temporal Processing Tasks 98%
- Mean-field approximations with adaptive coupling for networks with spike-timing-dependent plasticity 96%
- Efficient decoding of large-scale neural population responses with Gaussian-process multiclass regression 96%
Similar papers in this journal
- Spiking neural network models of sound localisation via a massively collaborative process 97%
- Synchrony in Networks of Type 2 Interneurons is More Robust to Noise with Hyperpolarizing Inhibition Compared to Shunting Inhibition in Both the Stochastic Population Oscillator and the Coupled Oscillator Regimes 96%
- Comparing surrogates to evaluate precisely timed higher-order spike correlations 95%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.