1 Supplementary Materials for : Emergence of complex computational structures from chaotic neural networks through reward-modulated Hebbian learning

By Gregor M. Hoerzer, Robert A. Legenstein, Wolfgang Maass. 2012

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Deep learning in neural networks: An overview (2014, background): “ Brunel, 2000; Bohte et al., 2002; Gerstner and Kistler, 2002; Izhikevich et al., 2003; Seung, 2003; Deco and Rolls, 2005; Brette et al., 2007; Brea et al., 2013; Nessler et al., 2013; Kasabov, 2014; Hoerzer et al., 2014; Rezende and Gerstner, 2014). Future energy-efﬁcient hardware for DL in NNs may”
Integration of nanoscale memristor synapses in neuromorphic computing architectures (2013, background): “informative reward signal, which just indicates whether some measure of performance of the system has recently increased [84].”
Backpropamine: training self-modifying neural networks with differentiable neuromodulated plasticity (2018, background): “Such mechanisms have been modelled in computational neuroscience studies, e.g. (Izhikevich, 2007; Hoerzer et al., 2014; Fiete et al., 2007; Soltoggio & Steil, 2013; Miconi, 2017) (see Gerstner et al. 2018 for a recent review of this concept).”
Interactive Control of Diverse Complex Characters with Neural Networks (2015, background): “The neural network weight matrices are initialized with a spectral radius of just above 1, similar to [15, 5].”
Recent Advances in Physical Reservoir Computing: A Review (2018, methodology): “…emerged working memory model through reward-modulated online learning rule that enables the same neural circuit to solve different tasks (Hoerzer, Legenstein, & Maass, 2012), and the comparative analysis of three working memory models including the RC-based one (Barak, Sussillo, Romo,…”
Motor primitives in space and time via targeted gain modulation in cortical networks (2018, unknown): “[12, 13]).”
RM-SORN: a reward-modulated self-organizing recurrent neural network (2015, background): “Another rate-based model with reward-modulated hebbian learning was created by Hoerzer et al. (2014).”
Neural Field Continuum Limits and the Structure–Function Partitioning of Cognitive–Emotional Brain Networks (2023, background): “These and additional field theory attributes of (non)random variation in neuronal activity and hidden layers reveal the organization of structurally ambiguous functional connectivity [94–96] and imply that some network partitioning, as opposed to unconstrained embeddedness, is required for the maxim”
Noise as a Resource for Computation and Learning in Networks of Spiking Neurons (2014, background): “, [61] and [62]), and it is only possible if the neurons or Maass: Noise as a Resource for Computation and Learning in Networks of Spiking Neurons”
Selection of cortical dynamics for motor behaviour by the basal ganglia (2015, background): “Recently, various works have highlighted that reservoir computing can be a candidate computational approach to describe the nature of cortical encoding (Wang 2008; Rigotti et al. 2010; Dominey 2013; Hoerzer et al. 2014).”