https://arxiv.org/abs/1610.06402 A Growing Long-term Episodic & Semantic Memory

To address this, we describe a lifelong learning system that leverages a fast, though non-differentiable, content-addressable memory which can be exploited to encode both a long history of sequential episodic knowledge and semantic knowledge over many episodes for an unbounded number of domains. This opens the door for investigation into transfer learning, and leveraging prior knowledge that has been learned over a lifetime of experiences to new domains.

Stretcher Networks Figure 2a shows a single stretcher network, that takes a relatively small vector and produces a much larger vector, which is then reshaped into the parameters of an LSTM auto-encoder. Figure 2b shows the same network used three times, producing three different LSTM autoencoders from three different “program” vectors. Note that the weights of the stretcher network are tied for all three instances.

https://arxiv.org/abs/1611.00020 Neural Symbolic Machines: Learning Semantic Parsers on Freebase with Weak Supervision

In this work, we propose the Manager-Programmer-Computer framework, which integrates neural networks with non-differentiable memory to support abstract, scalable and precise operations through a friendly neural computer interface. Specifically, we introduce a Neural Symbolic Machine, which contains a sequence-to-sequence neural “programmer”, and a non-differentiable “computer” that is a Lisp interpreter with code assist. To successfully apply REINFORCE for training, we augment it with approximate gold programs found by an iterative maximum likelihood training process. NSM is able to learn a semantic parser from weak supervision over a large knowledge base. It achieves new state-of-the-art performance on WebQuestionsSP, a challenging semantic parsing dataset, with weak supervision. Compared to previous approaches,

https://arxiv.org/pdf/1606.01994v2.pdf CFO: Conditional Focused Neural Question Answering with Large-scale Knowledge Bases

We propose CFO, a Conditional Focused neuralnetwork-based approach to answering factoid questions with knowledge bases. Our approach first zooms in a question to find more probable candidate subject mentions, and infers the final answers with a unified conditional probabilistic framework. Powered by deep recurrent neural networks and neural embeddings, our proposed CFO achieves an accuracy of 75.7% on a dataset of 108k questions – the largest public one to date. It outperforms the current state of the art by an absolute margin of 11.8%.

We employ a conditional factoid factorization by inferring the target relation first and then the target subject associated with the candidate relations. To resolve the representation for millions of entities, we proposed type-vector scheme which requires no training. Our focused pruning largely reduces the candidate space without loss of recall rate, leading to significant improvement of overall accuracy.

Overall structure of the subject network. Sigmoid layer is added only when type vector is used as E(s).

https://openreview.net/pdf?id=rJTKKKqeg TRACKING THE WORLD STATE WITH RECURRENT ENTITY NETWORKS

We introduce a new model, the Recurrent Entity Network (EntNet). It is equipped with a dynamic long-term memory which allows it to maintain and update a representation of the state of the world as it receives new data. For language understanding tasks, it can reason on-the-fly as it reads text, not just when it is required to answer a question or respond as is the case for a Memory Network. Like a Neural Turing Machine or Differentiable Neural Computer it maintains a fixed size memory and can learn to perform location and content-based read and write operations. However, unlike those models it has a simple parallel architecture in which several memory locations can be updated simultaneously. The EntNet sets a new state-of-the-art on the bAbI tasks, and is the first method to solve all the tasks in the 10k training examples setting. We also demonstrate that it can solve a reasoning task which requires a large number of supporting facts, which other methods are not able to solve, and can generalize past its training horizon. It can also be practically used on large scale datasets such as Children’s Book Test, where it obtains competitive performance, reading the story in a single pass.

https://arxiv.org/abs/1606.04422 Logic Tensor Networks: Deep Learning and Logical Reasoning from Data and Knowledge

https://arxiv.org/pdf/1612.01254v1.pdf Deep Symbolic Representation Learning for Heterogeneous Time-series Classification

In this paper, we consider the problem of event classification with multi-variate time series data consisting of heterogeneous (continuous and categorical) variables. The complex temporal dependencies between the variables combined with sparsity of the data makes the event classification problem particularly challenging. Most state-of-art approaches address this either by designing hand-engineered features or breaking up the problem over homogeneous variates. In this work, we propose and compare three representation learning algorithms over symbolized sequences which enables classification of heterogeneous time-series data using a deep architecture. The proposed representations are trained jointly along with the rest of the network architecture in an end-to-end fashion that makes the learned features discriminative for the given task. Experiments on three real-world datasets demonstrate the effectiveness of the proposed approaches.