A neural Turing machine (NTM) is a recurrent neural network model of a Turing machine. The approach was published by Alex Graves et al. in 2014. ^[1] NTMs combine the fuzzy pattern matching capabilities of neural networks with the algorithmic power of programmable computers.

An NTM has a neural network controller coupled to external memory resources, which it interacts with through attentional mechanisms. The memory interactions are differentiable end-to-end, making it possible to optimize them using gradient descent. ^[2] An NTM with a long short-term memory (LSTM) network controller can infer simple algorithms such as copying, sorting, and associative recall from examples alone. ^[1]

The authors of the original NTM paper did not publish their source code. ^[1] The first stable open-source implementation was published in 2018 at the 27th International Conference on Artificial Neural Networks, receiving a best-paper award. ^{[3] [4] [5]} Other open source implementations of NTMs exist but as of 2018 they are not sufficiently stable for production use. ^{[6] [7] [8] [9] [10] [11] [12]} The developers either report that the gradients of their implementation sometimes become NaN during training for unknown reasons and cause training to fail; ^{[10] [11] [9]} report slow convergence; ^{[7] [6]} or do not report the speed of learning of their implementation. ^{[12] [8]}

Differentiable neural computers are an outgrowth of Neural Turing machines, with attention mechanisms that control where the memory is active, and improve performance. ^[13]

References