Competitive Learning: From Interactive Activation to Adaptive Resonance

By Stephen Grossberg. 1987

Subfields: Representation & Self-Supervised Learning

How researchers see this paper

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

In Representation & Self-Supervised Learning

Citations in the corpus

Lifelong learning of human actions with deep neural network self-organization (2017, background): “When dealing with non-stationary data, processing systems should account for the integration of new information while preserving existing knowledge (Grossberg, 1987).”
Structural Synaptic Plasticity Has High Memory Capacity and Can Explain Graded Amnesia, Catastrophic Forgetting, and the Spacing Effect (2014, background): “ing catastrophic forgetting in brains [36,37], graded retrograde”
Generalization Through the Recurrent Interaction of Episodic Memories (2012, methodology): “These IAC networks, which draw on related ideas of Grossberg (1978; see also Grossberg, 1987), have been widely used in the past to characterize cognitive processes involved in perception and memory (McClelland, 1981; McClelland & Rumelhart, 1981; Nystrom & McClelland, 1992)”
Dendritic cortical microcircuits approximate the backpropagation algorithm (2018, background): “Interestingly, the workhorse of deep learning is still the classical backpropagation of errors algorithm (backprop; Rumelhart et al., 1986), which has been long dismissed in neuro-science on the grounds of biologically implausibility (Grossberg, 1987; Crick, 1989).”
Assessing the Scalability of Biologically-Motivated Deep Learning Algorithms and Architectures (2018, background): “Historically, there has been significant disagreement about whether BP can tell us anything interesting about learning in the brain [8, 11].”
Locating object knowledge in the brain: comment on Bowers's (2009) attempt to revive the grandmother cell hypothesis. (2010, background): “A localist theorist would likely claim that only some of these distinctions should be reified by assigning a localist unit to each of the alternative subvarieties, but wherever the localist stops (whether manually or on the basis of a vigilance parameter; Grossberg, 1987), a problem will remain: The”
The Visual System's Internal Model of the World (2015, background): “Five major classes of computational models have been proposed over the last 40 years on how the hierarchical internal models in the visual cortex are constructed and function to support perceptual learning and inference.”
The Life-Course Dynamics of Goal Pursuit and Goal Adjustment: A Two-Process Framework (2002, background): “This stability–flexibility dilemma (Bak & Brandtstädter, 1998; see also Grossberg, 1987) also arises in contexts of life planning in which the choice and implementation of goals depends on resources and constraints that change on ontogenetic and historical levels.”
A Dual-Route Approach to Orthographic Processing (2011, methodology): “Unsupervised learning algorithms, such as implemented in Self-Organizing Maps (e.g., Kohonen, 1982; Dufau et al., 2010) or Adaptive Resonance Theory (e.g., Grossberg, 1987; Glotin et al., 2010), provide specific mechanisms for performing orthographic chunking on the basis of a parallel letter input.”
Towards deep learning with segregated dendrites (2016, background): “Perhaps the most biologically unrealistic component of backpropagation is the use of non-local “weight-transport” (Figure 1B) (Grossberg, 1987).”
Forward Models: Supervised Learning with a Distal Teacher (1992, background): “The second assumption has also been weakened: Learning algorithms that require no explicit teacher have been developed (Becker & Hinton, 1989; Grossberg, 1987; Kohonen, 1982; Linsker, 1988; Rumelhart & Zipser, 1986).”
Random feedback weights support learning in deep neural networks (2014, background): “But the most powerful class of algorithms use error signals tailored to each neuron and seemed impossible to implement in the brain because they required weight transport 6,7 .”