Deep neural networks show great potential as solutions to many sensing application problems, but their excessive resource demand slows down execution time, pausing a serious impediment to deployment on low-end devices. To address this challenge, recent literature focused on compressing neural network size to improve performance. We show that changing neural network size does not proportionally affect performance attributes of interest, such as execution time. Rather, extreme run-time nonlinearities exist over the network configuration space. Hence, we propose a novel framework, called FastDeepIoT, that uncovers the non-linear relation between neural network structure and execution time, then exploits that understanding to find network configurations that significantly improve the trade-off between execution time and accuracy on mobile and embedded devices. FastDeepIoT makes two key contributions. First, FastDeepIoT automatically learns an accurate and highly interpretable execution time model for deep neural networks on the target device. This is done without prior knowledge of either the hardware specifications or the detailed implementation of the used deep learning library. Second, FastDeepIoT informs a compression algorithm how to minimize execution time on the profiled device without impacting accuracy. We evaluate FastDeepIoT using three different sensing-related tasks on two mobile devices: Nexus 5 and Galaxy Nexus. FastDeepIoT further reduces the neural network execution time by $48\%$ to $78\%$ and energy consumption by $37\%$ to $69\%$ compared with the state-of-the-art compression algorithms.

Continuous word representation (aka word embedding) is a basic building block in many neural network-based models used in natural language processing tasks. Although it is widely accepted that words with similar semantics should be close to each other in the embedding space, we find that word embeddings learned in several tasks are biased towards word frequency: the embeddings of high-frequency and low-frequency words lie in different subregions of the embedding space, and the embedding of a rare word and a popular word can be far from each other even if they are semantically similar. This makes learned word embeddings ineffective, especially for rare words, and consequently limits the performance of these neural network models. In this paper, we develop a neat, simple yet effective way to learn \emph{FRequency-AGnostic word Embedding} (FRAGE) using adversarial training. We conducted comprehensive studies on ten datasets across four natural language processing tasks, including word similarity, language modeling, machine translation and text classification. Results show that with FRAGE, we achieve higher performance than the baselines in all tasks.

We estimate the proper channel (width) scaling of Convolution Neural Networks (CNNs) for model reduction. Unlike the traditional scaling method that reduces every CNN channel width by the same scaling factor, we address each CNN macroblock adaptively depending on its information redundancy measured by our proposed effective flops. Our proposed macroblock scaling (MBS) algorithm can be applied to various CNN architectures to reduce their model size. These applicable models range from compact CNN models such as MobileNet (25.53% reduction, ImageNet) and ShuffleNet (20.74% reduction, ImageNet) to ultra-deep ones such as ResNet-101 (51.67% reduction, ImageNet) and ResNet-1202 (72.71% reduction, CIFAR-10) with negligible accuracy degradation. MBS also performs better reduction at a much lower cost than does the state-of-the-art optimization-based method. MBS's simplicity and efficiency, its flexibility to work with any CNN model, and its scalability to work with models of any depth makes it an attractive choice for CNN model size reduction.

Semantic frame parsing is a crucial component in spoken language understanding (SLU) to build spoken dialog systems. It has two main tasks: intent detection and slot filling. Although state-of-the-art approaches showed good results, they require large annotated training data and long training time. In this paper, we aim to alleviate these drawbacks for semantic frame parsing by utilizing the ubiquitous user information. We design a novel coarse-to-fine deep neural network model to incorporate prior knowledge of user information intermediately to better and quickly train a semantic frame parser. Due to the lack of benchmark dataset with real user information, we synthesize the simplest type of user information (location and time) on ATIS benchmark data. The results show that our approach leverages such simple user information to outperform state-of-the-art approaches by 0.25% for intent detection and 0.31% for slot filling using standard training data. When using smaller training data, the performance improvement on intent detection and slot filling reaches up to 1.35% and 1.20% respectively. We also show that our approach can achieve similar performance as state-of-the-art approaches by using less than 80% annotated training data. Moreover, the training time to achieve the similar performance is also reduced by over 60%.

While a lot of progress has been made in recent years, the dynamics of learning in deep nonlinear neural networks remain to this day largely misunderstood. In this work, we study the case of binary classification and prove various properties of learning in such networks under strong assumptions such as linear separability of the data. Extending existing results from the linear case, we confirm empirical observations by proving that the classification error also follows a sigmoidal shape in nonlinear architectures. We show that given proper initialization, learning expounds parallel independent modes and that certain regions of parameter space might lead to failed training. We also demonstrate that input norm and features' frequency in the dataset lead to distinct convergence speeds which might shed some light on the generalization capabilities of deep neural networks. We provide a comparison between the dynamics of learning with cross-entropy and hinge losses, which could prove useful to understand recent progress in the training of generative adversarial networks. Finally, we identify a phenomenon that we baptize gradient starvation where the most frequent features in a dataset prevent the learning of other less frequent but equally informative features.

Automatic judgment prediction aims to predict the judicial results based on case materials. It has been studied for several decades mainly by lawyers and judges, considered as a novel and prospective application of artificial intelligence techniques in the legal field. Most existing methods follow the text classification framework, which fails to model the complex interactions among complementary case materials. To address this issue, we formalize the task as Legal Reading Comprehension according to the legal scenario. Following the working protocol of human judges, LRC predicts the final judgment results based on three types of information, including fact description, plaintiffs' pleas, and law articles. Moreover, we propose a novel LRC model, AutoJudge, which captures the complex semantic interactions among facts, pleas, and laws. In experiments, we construct a real-world civil case dataset for LRC. Experimental results on this dataset demonstrate that our model achieves significant improvement over state-of-the-art models. We will publish all source codes and datasets of this work on \urlgithub.com for further research.

With joint learning of sampling and recovery, the deep learning-based compressive sensing (DCS) has shown significant improvement in performance and running time reduction. Its reconstructed image, however, losses high-frequency content especially at low subrates. This happens similarly in the multi-scale sampling scheme which also samples more low-frequency components. In this paper, we propose a multi-scale DCS convolutional neural network (MS-DCSNet) in which we convert image signal using multiple scale-based wavelet transform, then capture it through convolution block by block across scales. The initial reconstructed image is directly recovered from multi-scale measurements. Multi-scale wavelet convolution is utilized to enhance the final reconstruction quality. The network is able to learn both multi-scale sampling and multi-scale reconstruction, thus results in better reconstruction quality.

"I'm sorry, I didn't quite get that." Anyone who has ever tried to have a deeper conversation with a virtual assistant like Siri knows how frustrating it can be. That's despite the fact that AI systems like it are increasingly pushing into our lives, with new success stories on an almost daily basis. Not only do AIs now help radiologists detect tumours, they can act as cat repellent and even detect signals of potential alien technology from space. But when it comes to fundamental human abilities, like having a good chat, AI falls short.

In this post we are going to develop a java face recognition application using deeplearning4j. The application is offering a GUI and flexibility to register new faces so feel free to try with your own images. Additionally you can check out the free open source code as part of the PactPub video course Java Machine Learning for Computer Vision together with many new improvements to previous posts applications in java. Face recognition has always been an important problem to solve due its sensitivity in regards to security and because it closely related to people identity. For many years face recognition applications were well known especially in criminology and searching for wanted persons with cameras and sometimes even using satellites. Now days, in deep-learning era face recognition is widely found from simple applications to unlocking your phone offering state of the art accuracy.

As you can see, there is massive overlap between these generative sequence models and AEs we discussed previously. A very popular example of sequence modelling is their application to NLP or text modelling.