Our project focus on the problem of aspect specific sentiment analysis using recursive neural networks. Different from the previous studies where labels exist on every node of constituency tree, we have only one label each sentence, which is only on the root node, and it causes a severe vanishing gradient problem for both RNN and RNTN. To deal with such problem, we develop a classification algo...

In recent years, gradient-based LSTM recurrent neural networks (RNNs) solved many previously RNN-unlearnable tasks. Sometimes, however, gradient information is of little use for training RNNs, due to numerous local minima. For such cases, we present a novel method: EVOlution of systems with LINear Outputs (Evolino). Evolino evolves weights to the nonlinear, hidden nodes of RNNs while computing ...

In this paper, we propose a novel neural approach for paraphrase generation. Conventional paraphrase generation methods either leverage hand-written rules and thesauri-based alignments, or use statistical machine learning principles. To the best of our knowledge, this work is the first to explore deep learning models for paraphrase generation. Our primary contribution is a stacked residual LSTM...

We show that a deep convolutional network with an architecture inspired by the models used in image recognition can yield accuracy similar to a long-short term memory (LSTM) network, which achieves the state-of-the-art performance on the standard Switchboard automatic speech recognition task. Moreover, we demonstrate that merging the knowledge in the CNN and LSTM models via model compression fu...

We explored two approaches based on Long Short-Term Memory (LSTM) networks on the Quora duplicate question dataset. The first model uses a Siamese architecture with the learned representations from a single LSTM running on both sentences. The second method uses two LSTMs with the two sentences in sequence, and the second attending on the first (word-by-word attention). Our best model achieved 7...

Selecting appropriate words to compose a sentence is one common problem faced by non-native Chinese learners. In this paper, we propose (bidirectional) LSTM sequence labeling models and explore various features to detect word usage errors in Chinese sentences. By combining CWINDOW word embedding features and POS information, the best bidirectional LSTM model achieves accuracy 0.5138 and MRR 0.6...

While much work has been done on unsupervised learning in feedforward neural network architectures, its potential with (theoretically more powerful) recurrent networks and time-varying inputs has rarely been explored. Here we train Long Short-Term Memory (LSTM) recurrent networks to maximize two information-theoretic objectives for unsupervised learning: Binary Information Gain Optimization (BI...

This study compares the sequential and parallel efficiency of training Recurrent Neural Networks (RNNs) with Hessian-free optimization versus a gradient descent variant. Experiments are performed using the long short term memory (LSTM) architecture and the newly proposed multiplicative LSTM (mLSTM) architecture. Results demonstrate a number of insights into these architectures and optimization ...

Abstract—This paper analyzes fundamental ideas and concepts related to neural networks, which provide the reader a theoretical explanation of Long Short-Term Memory (LSTM) networks operation classified as Deep Learning Systems, and to explicitly present the mathematical development of Backward Pass equations of the LSTM network model. This mathematical modeling associated with software developm...

We show that collaborative filtering can be viewed as a sequence prediction problem, and that given this interpretation, recurrent neural networks offer very competitive approach. In particular we study how the long short-term memory (LSTM) can be applied to collaborative filtering, and how it compares to standard nearest neighbors and matrix factorization methods on movie recommendation. We sh...