Abstract / Bemerkung
Memory-augmented neural networks equip a recurrent neural network with an explicit memory to support tasks that require information storage without interference over long times. A key motivation for such research is to perform classic computation tasks, such as parsing. However, memory-augmented neural networks are notoriously hard to train, requiring many backpropagation epochs and a lot of data. In this paper, we introduce the reservoir stack machine, a model which can provably recognize all deterministic context-free languages and circumvents the training problem by training only the output layer of a recurrent net and employing auxiliary information during training about the desired interaction with a stack. In our experiments, we validate the reservoir stack machine against deep and shallow networks from the literature on three benchmark tasks for Neural Turing machines and six deterministic context-free languages. Our results show that the reservoir stack machine achieves zero error, even on test sequences longer than the training data, requiring only a few seconds of training time and 100 training sequences.
reservoir computing; neural turing machines; pushdown automata; context-free grammars; echo state networks
Paaßen B, Schulz A, Hammer B. Reservoir Stack Machines. Neurocomputing. Accepted.
Paaßen, B., Schulz, A., & Hammer, B. (Accepted). Reservoir Stack Machines. Neurocomputing
Paaßen, B., Schulz, A., and Hammer, B. (Accepted). Reservoir Stack Machines. Neurocomputing.
Paaßen, B., Schulz, A., & Hammer, B., Accepted. Reservoir Stack Machines. Neurocomputing.
B. Paaßen, A. Schulz, and B. Hammer, “Reservoir Stack Machines”, Neurocomputing, Accepted.
Paaßen, B., Schulz, A., Hammer, B.: Reservoir Stack Machines. Neurocomputing. (Accepted).
Paaßen, Benjamin, Schulz, Alexander, and Hammer, Barbara. “Reservoir Stack Machines”. Neurocomputing (Accepted).
Link(s) zu Volltext(en)
Software implementation and experimental code