This is the second of three blogs on the state of the AI chip market and what's to come in 2019. The year will be a festival of new chips and benchmark battles, led by the large companies I mentioned in the first blog ( Intel, Google, AMD, Xilinx, Apple, Qualcomm), and joined by dozens of Silicon Valley startups and Chinese Unicorns sporting valuations in excess of a billion US dollars. In this section, I will cover the most prominent, or at least the loudest, of the startups in the West and in China, where the government is intent on creating an indigenous AI chip industry. We will start with Wave, which appears to be the first to market with silicon for training. Wave Computing had an eventful 2018, taping out its first DataFlow Processing Unit, acquiring MIPS, creating MIPS Open, and shipping its first early systems to a few lucky customers.

Embattled Chinese telecom giant Huawei has some new problems. The US Department of Justice on Monday unsealed a 13-count indictment against Huawei and its CFO, Meng Wanzhou, alleging the company misled banking partners about violations of US sanctions against Iran. The charges include bank fraud, wire fraud, money laundering, and obstruction of justice. Meng, who is also the daughter of Huawei founder Ren Zhengfei, was arrested in Canada last month and is awaiting extradition to the US. In a separate case, the DOJ indicted Huawei for stealing intellectual property related to a cell-phone-testing robot from T-Mobile in 2012.

YOKOHAMA - Toshiba Corp. unveiled a remote-controlled robot with tongs on Monday that it hopes will be able to probe the inside of one of the three damaged reactors at Japan's tsunami-hit Fukushima nuclear plant and grip chunks of highly radioactive melted fuel. The device is designed to slide down an extendable 11-meter (36-foot) long pipe and touch melted fuel inside reactor 2's primary containment vessel. The reactor was built by Toshiba and GE. An earlier probe carrying a camera captured images of pieces of melted fuel in the reactor last year, and robotic probes in the two other reactors have detected traces of damaged fuel, but the exact location, contents and other details remain largely unknown. Toshiba's energy systems unit said experiments with the new probe planned in February are key to determining the proper equipment and technologies needed to remove the fuel debris, the most challenging part of the decommissioning process expected to take decades.

Advances in robotics, artificial intelligence, and machine learning are ushering in a new age of automation, as machines match or outperform human performance. Machine intelligence can enable businesses to improve performance by reducing errors, improving sensitivity, quality and speed, and in some cases achieving outcomes that go beyond current resource capabilities. Relevant applications include new product architecture design, rapid material characterization, and life-cycle management tied with a digital strategy that will enable efficient development of products from cradle to grave. In addition, there are also challenges to overcome that must be addressed through a major, sustained research effort that is based solidly on both inferential and computational principles applied to design tailoring of functionally optimized structures. Current applications of structural materials in the aerospace industry demand the highest quality control of material microstructure, especially for advanced rotational turbomachinery in aircraft engines in order to have the best tailored material property. In this paper, deep convolutional neural networks were developed to accurately predict processing-structure-property relations from materials microstructures images, surpassing current best practices and modeling efforts. The models automatically learn critical features, without the need for manual specification and/or subjective and expensive image analysis. Further, in combination with generative deep learning models, a framework is proposed to enable rapid material design space exploration and property identification and optimization. The implementation must take account of real-time decision cycles and the trade-offs between speed and accuracy.

Citizen engagement and technology usage are two emerging trends driven by smart city initiatives. Governments around the world are adopting technology for faster resolution of civic issues. Typically, citizens report issues, such as broken roads, garbage dumps, etc. through web portals and mobile apps, in order for the government authorities to take appropriate actions. Several mediums -- text, image, audio, video -- are used to report these issues. Through a user study with 13 citizens and 3 authorities, we found that image is the most preferred medium to report civic issues. However, analyzing civic issue related images is challenging for the authorities as it requires manual effort. Moreover, previous works have been limited to identifying a specific set of issues from images. In this work, given an image, we propose to generate a Civic Issue Graph consisting of a set of objects and the semantic relations between them, which are representative of the underlying civic issue. We also release two multi-modal (text and images) datasets, that can help in further analysis of civic issues from images. We present a novel approach for adversarial training of existing scene graph models that enables the use of scene graphs for new applications in the absence of any labelled training data. We conduct several experiments to analyze the efficacy of our approach, and using human evaluation, we establish the appropriateness of our model at representing different civic issues.

Taxonomies are important building blocks of structured knowledge bases, and their construction from text sources and Wikipedia has received much attention. In this paper we focus on the construction of taxonomies for fictional domains, using noisy category systems from fan wikis or text extraction as input. Such fictional domains are archetypes of entity universes that are poorly covered by Wikipedia, such as also enterprise-specific knowledge bases or highly specialized verticals. Our fiction-targeted approach, called TiFi, consists of three phases: (i) category cleaning, by identifying candidate categories that truly represent classes in the domain of interest, (ii) edge cleaning, by selecting subcategory relationships that correspond to class subsumption, and (iii) top-level construction, by mapping classes onto a subset of high-level WordNet categories. A comprehensive evaluation shows that TiFi is able to construct taxonomies for a diverse range of fictional domains such as Lord of the Rings, The Simpsons or Greek Mythology with very high precision and that it outperforms state-of-the-art baselines for taxonomy induction by a substantial margin.

A model used for velocity control during car following was proposed based on deep reinforcement learning (RL). To fulfil the multi-objectives of car following, a reward function reflecting driving safety, efficiency, and comfort was constructed. With the reward function, the RL agent learns to control vehicle speed in a fashion that maximizes cumulative rewards, through trials and errors in the simulation environment. A total of 1,341 car-following events extracted from the Next Generation Simulation (NGSIM) dataset were used to train the model. Car-following behavior produced by the model were compared with that observed in the empirical NGSIM data, to demonstrate the model's ability to follow a lead vehicle safely, efficiently, and comfortably. Results show that the model demonstrates the capability of safe, efficient, and comfortable velocity control in that it 1) has small percentages (8\%) of dangerous minimum time to collision values (\textless\ 5s) than human drivers in the NGSIM data (35\%); 2) can maintain efficient and safe headways in the range of 1s to 2s; and 3) can follow the lead vehicle comfortably with smooth acceleration. The results indicate that reinforcement learning methods could contribute to the development of autonomous driving systems.

Short-term traffic flow prediction is one of the crucial issues in intelligent transportation system, which is an important part of smart cities. Accurate predictions can enable both the drivers and the passengers to make better decisions about their travel route, departure time and travel origin selection, which can be helpful in traffic management. Multiple models and algorithms based on time series prediction and machine learning were applied to this issue and achieved acceptable results. Recently, the availability of sufficient data and computational power, motivates us to improve the prediction accuracy via deep-learning approaches. Recurrent neural networks have become one of the most popular methods for time series forecasting, however, due to the variety of these networks, the question that which type is the most appropriate one for this task remains unsolved. In this paper, we use three kinds of recurrent neural networks including simple RNN units, GRU and LSTM neural network to predict short-term traffic flow. The dataset from TAP30 Corporation is used for building the models and comparing RNNs with several well-known models, such as DEMA, LASSO and XGBoost. The results show that all three types of RNNs outperform the others, however, more simple RNNs such as simple recurrent units and GRU perform work better than LSTM in terms of accuracy and training time.

Computations for the softmax function are significantly expensive when the number of output classes is large. In this paper, we present a novel softmax inference speedup method, Doubly Sparse Softmax (DS-Softmax), that leverages sparse mixture of sparse experts to efficiently retrieve top-k classes. Different from most existing methods that require and approximate a fixed softmax, our method is learning-based and can adapt softmax weights for a better approximation. In particular, our method learns a two-level hierarchy which divides entire output class space into several partially overlapping experts. Each expert is sparse and only contains a subset of output classes. To find top-k classes, a sparse mixture enables us to find the most probable expert quickly, and the sparse expert enables us to search within a small-scale softmax. We empirically conduct evaluation on several real-world tasks (including neural machine translation, language modeling and image classification) and demonstrate that significant computation reductions can be achieved without loss of performance.

We investigate the problem of multiclass classification with rejection, where a classifier can choose not to make a prediction to avoid critical misclassification. We consider two approaches for this problem: a traditional one based on confidence scores and a more recent one based on simultaneous training of a classifier and a rejector. An existing method in the former approach focuses on a specific class of losses and its empirical performance is not very convincing. In this paper, we propose confidence-based rejection criteria for multiclass classification, which can handle more general losses and guarantee calibration to the Bayes-optimal solution. The latter approach is relatively new and has been available only for the binary case, to the best of our knowledge. Our second contribution is to prove that calibration to the Bayes-optimal solution is almost impossible by this approach in the multiclass case. Finally, we conduct experiments to validate the relevance of our theoretical findings.