Unsupervised learning by competing hidden units

By Dmitry Krotov, John J. Hopfield. 2018

187 citations recorded by the source corpus.

Significance Despite great success of deep learning a question remains to what extent the computational properties of deep neural networks are similar to those of the human brain. The particularly nonbiological aspect of deep learning is the supervised training process with the backpropagation algorithm, which requires massive amounts of labeled data, and a nonlocal learning rule for changing the synapse strengths. This paper describes a learning algorithm that does not suffer from these two problems. It learns the weights of the lower layer of neural networks in a completely unsupervised fashion. The entire algorithm utilizes local learning rules which have conceptual biological plausibility. It is widely believed that end-to-end training with the backpropagation algorithm is essential for learning good feature detectors in early layers of artificial neural networks, so that these detectors are useful for the task performed by the higher layers of that neural network. At the same time, the traditional form of backpropagation is biologically implausible. In the present paper we propose an unusual learning rule, which has a degree of biological plausibility and which is motivated by Hebb’s idea that change of the synapse strength should be local—i.e., should depend only on the activities of the pre- and postsynaptic neurons. We design a learning algorithm that utilizes global inhibition in the hidden layer and is capable of learning early feature detectors in a completely unsupervised way. These learned lower-layer feature detectors can be used to train higher-layer weights in a usual supervised way so that the performance of the full network is comparable to the performance of standard feedforward networks trained end-to-end with a backpropagation algorithm on simple tasks.

Subfields: Artificial Intelligence, Electrical and Electronic Engineering, 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 Artificial Intelligence

1. Song-Chun Zhu
2. Erik Nijkamp
3. Ying Nian Wu
4. Mitch Hill
5. Quanshi Zhang
6. Mengzhou Li
7. Jinjun Xiong
8. Ge Wang

In Electrical and Electronic Engineering

1. Song-Chun Zhu
2. Erik Nijkamp
3. Ying Nian Wu

In Representation & Self-Supervised Learning

1. Song-Chun Zhu
2. Ying Nian Wu
3. Erik Nijkamp
4. Mitch Hill
5. Quanshi Zhang
6. Jianwen Xie
7. Yang Lu

Citations in the corpus

• Learning Non-Convergent Non-Persistent Short-Run MCMC Toward Energy-Based Model (2019, background): “This appears to be consistent with the common practice of computational neuroscience [27], where researchers often directly start from the dynamics, such as attractor dynamics [20, 2, 36] whose express goal is to 2”
• On Interpretability of Artificial Neural Networks: A Survey (2020, background): “For example, in a biological neural system, synapses are updated in a local manner [94] and only depend on the activities of presynaptic and postsynaptic neurons.”