Memory and attention in deep learning

Abstract

Intelligence necessitates memory. Without memory, humans fail to perform various nontrivial tasks such as reading novels, playing games or solving maths. As the ultimate goal of machine learning is to derive intelligent systems that learn and act automatically just like human, memory construction for machine is inevitable. Artificial neural networks model neurons and synapses in the brain by interconnecting computational units via weights, which is a typical class of machine learning algorithms that resembles memory structure. Their descendants with more complicated modeling techniques (a.k.a deep learning) have been successfully applied to many practical problems and demonstrated the importance of memory in the learning process of machinery systems. Recent progresses on modeling memory in deep learning have revolved around external memory constructions, which are highly inspired by computational Turing models and biological neuronal systems. Attention mechanisms are derived to support acquisition and retention operations on the external memory. Despite the lack of theoretical foundations, these approaches have shown promises to help machinery systems reach a higher level of intelligence. The aim of this thesis is to advance the understanding on memory and attention in deep learning. Its contributions include: (i) presenting a collection of taxonomies for memory, (ii) constructing new memory-augmented neural networks (MANNs) that support multiple control and memory units, (iii) introducing variability via memory in sequential generative models, (iv) searching for optimal writing operations to maximise the memorisation capacity in slot-based memory networks, and (v) simulating the Universal Turing Machine via Neural Stored-program Memory-a new kind of external memory for neural networks.

PHD Thesis

References in corpus (18)

• Sequence to Sequence Learning with Neural Networks
• Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
• On the difficulty of training Recurrent Neural Networks
• WaveNet: A Generative Model for Raw Audio
• Semi-Supervised Learning with Deep Generative Models
• On the Properties of Neural Machine Translation: Encoder-Decoder Approaches
• Understanding deep learning requires rethinking generalization
• A Survey on Multi-view Learning
• A Simple Way to Initialize Recurrent Networks of Rectified Linear Units
• Deep Unsupervised Clustering with Gaussian Mixture Variational Autoencoders
• A Closer Look at Memorization in Deep Networks
• Using Fast Weights to Attend to the Recent Past
• Full-Capacity Unitary Recurrent Neural Networks
• Improving Neural Language Models with a Continuous Cache
• Memory Augmented Neural Networks with Wormhole Connections
• Generative Temporal Models with Memory
• Improving Long Handwritten Text Line Recognition with Convolutional Multi-way Associative Memory
• Neurocoder: Learning General-Purpose Computation Using Stored Neural Programs