Evolving Modular Fast-Weight Networks for Control

By Faustino J. Gomez, J. Urgen Schmidhuber. 2005

50 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Theory and Mathematics, Reinforcement Learning

How researchers see this paper

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

In Artificial Intelligence

In Computational Theory and Mathematics

In Reinforcement Learning

Citations in the corpus

HyperNetworks (2016, background): “A subsequent work demonstrated practical applications of fast weights (Gomez & Schmidhuber, 2005), where a generator network is learnt through evolution to solve an artiﬁcial control problem.”
Dynamic Filter Networks (2016, positive): “Our work also shares similar ideas with early work on fast-weight networks [4], that is, having a network learn to generate context dependent weights for another network.”
Meta Networks (2017, methodology): “The approach proposed by Gomez & Schmidhuber (2005) is more closely related to our work.”
A Brief Survey of Deep Reinforcement Learning (2017, positive): “eschewed the commonly used backpropagation algorithm in favour of evolutionary algorithms [32, 19, 53], which are gradient-free policy search algorithms.”
Boosting Few-Shot Visual Learning With Self-Supervision (2019, neutral): “, training images, of a few-shot task [26, 33, 53, 55, 57, 59]; methods learning to map a test example to a class label by accessing memory modules that store training examples for that task [11, 25, 35, 38, 50]; approaches that learn how to generate the weights of a classifier [12, 16, 43, 44, 13] ”
Training Agents using Upside-Down Reinforcement Learning (2019, methodology): “For example, Gomez and Schmidhuber [2005] trained non-linear recurrent networks to generate weights for linear controllers that could switch between multiple modes of control.”
Auto-Agent-Distiller: Towards Efficient Deep Reinforcement Learning Agents via Neural Architecture Search (2020, methodology): “Value-based methods [78, 57] aim to build a value function, which subsequently allows the deﬁnition of a policy; and policy-based methods [67] directly model the policy to maximize the expected return in either a gradient-free [23, 13] or gradient-based way [67].”
Evolutionary Computation for Reinforcement Learning (2012, methodology): “…context, this internal state can record aspects of the agent's observation history , which can help it cope with partial observability (Wieland, 1991; Gomez and Miikkulainen, 1999; Moriarty et al, 1999; Stanley and Miikkulainen, 2002; Igel, 2003; Gomez and Schmidhuber, 2005a,b).”
Training Recurrent Networks by Evolino (2007, background): “This approach can quickly learn t o solve difficult reinforcement learning control tasks [5,6,22], including ones that require use of deep memory [7].”
Online Bag-of-Visual-Words Generation for Unsupervised Representation Learning (2020, background): “We highlight that dynamic weight-generation modules are extensively used in the context of few-shot learning for producing classification weight vectors of novel classes using as input a limited set of training examples [27, 31, 58].”
Diet Networks: Thin Parameters for Fat Genomic (2016, unknown): “The idea of having two networks interacting with each other and with one producing parameters for the other is well rooted in the machine learning literature (Bengio et al., 1991; Schmidhuber, 1992; Gomez & Schmidhuber, 2005; Stanley et al., 2009; Denil et al., 2013; Andrychowicz et al., 2016).”
OBoW: Online Bag-of-Visual-Words Generation for Self-Supervised Learning (2020, background): “We highlight that dynamic weight-generation modules are extensively used in the context of few-shot learning for producing classification weight vectors of novel classes using as input a limited set of training examples [26, 30, 55].”