Are neural networks biologically plausible?

Training deep neural networks with the error backpropagation algorithm is considered implausible from a biological perspective. These spiking models achieve 98.2\% test accuracy on MNIST, which is close to the performance of rate networks with one hidden layer trained with backpropagation.

How biologically plausible are artificial neural networks?

Abstract: Artificial neural networks (ANNs) lack in biological plausibility, chiefly because backpropagation requires a variant of plasticity (precise changes of the synaptic weights informed by neural events that occur downstream in the neural circuit) that is profoundly incompatible with the current understanding of …

What is meant by biological plausibility?

In epidemiology and biomedicine, biological plausibility is the proposal of a causal association — a relationship between a putative cause and an outcome — that is consistent with existing biological and medical knowledge.

Why is backpropagation not biologically plausible?

Beyond Back-Prop CNNs aren’t biologically plausible due to their reliance on weights being exactly equal across multiple locations. RNNs rely on Back-Propagation Through Time (BPTT), of which there is currently no biologically plausible analogue [2].

What is the other name for RNN?

Feedback Neural Network
2.2 Recurrent or Feedback Neural Network RNN or feedback neural network is the second kind of ANN model, in which the outputs from neurons are used as feedback to the neurons of the previous layer. In other words, the current output is considered as an input for the next output.

Why is biological plausibility important?

Biological plausibility is an important criterion in evidence based decision making especially in public health because our evidence usually comes from observational rather than controlled studies.

What does the word plausible?

Today the word plausible usually means “reasonable” or “believable,” but it once held the meanings “worthy of being applauded” and “approving.” It comes to us from the Latin adjective plausibilis (“worthy of applause”), which in turn derives from the verb plaudere, meaning “to applaud or clap.” Other “plaudere” …

Does the brain use backpropagation?

Backprop in the brain? There is no direct evidence that the brain uses a backprop-like algorithm for learning. Past work has shown, however, that backprop-trained models can account for observed neural responses, such as the response properties of neurons in the posterior parietal cortex68 and primary motor cortex69.