Judgment prediction for legal cases has attracted much research efforts for its practice use, of which the ultimate goal is prison term prediction. While existing work merely predicts the total prison term, in reality a defendant is often charged with multiple crimes. In this paper, we argue that charge-based prison term prediction (CPTP) not only better fits realistic needs, but also makes the total prison term prediction more accurate and interpretable. We collect the first large-scale structured data for CPTP and evaluate several competitive baselines. Based on the observation that fine-grained feature selection is the key to achieving good performance, we propose the Deep Gating Network (DGN) for charge-specific feature selection and aggregation. Experiments show that DGN achieves the state-of-the-art performance.

For text classification in particular, most of the existing explanation methods identify parts of the input text which contribute most towards the predicted class (so called attribution methods or relevance methods) by exploiting various techniques such as input perturbation (Li et al., 2016), gradient analysis (Dimopoulos et al., 1995), and relevance propagation (Arras et al., 2017b). Besides, there are other explanation methods designed for specific deep learning architectures such as attention mechanism (Ghaeini et al., 2018) and extrac-tive rationale generation (Lei et al., 2016). We select some well-known explanation methods (which are applicable to CNNs for text classification) and evaluate them together with two new explanation methods proposed in this paper.

Many researchers have shown that the accuracy of deep learning models improves with larger data sets. The improvement is so significant that researchers often use sophisticated data augmentation techniques to synthetically generate extra data for model training. For example, this data set from ImageNet used to benchmark various deep learning image classification algorithms contains more than 14 million images with a million annotations. While the ResNet-50 model often used to benchmark image classification hardware is just over 100 MB in size. The models, which ideally are held in memory, must be continually fed with data and that often results in the storage system becoming the bottleneck to overall performance.

The rise of deep-learning (DL) has been fueled by the improvements in accelerators. Accelerators allow DL models to crunch a large amount of data, which is vital for them to achieve high accuracy. In fact, AlexNet, the famous winner of the ILSVRC 2012 competition, was trained on GPUs. GPU continues to remain the most widely used accelerator for DL applications, due to several of its features, such as high performance, continued improvements in its architecture and software-stack, ease of programming using high-level languages such as CUDA and availability of GPUs in cloud. "Accelerating DL models" is chasing a moving target As DL models are becoming more pervasive and accurate, their compute and memory requirements are growing tremendously.

In the next 5 years robotics and automation will streamline the supply chain in ways once thought impossible. Artificial Intelligence is a game changer. "Artificial Intelligence, deep learning, machine learning -- whatever you're doing if you don't understand it -- learn it. Because otherwise you're going to be a dinosaur within 3 years." "In the next 5 years we will see supply chain robotics and automation turn industries upside down."

Most people can identify objects in the picture, and what is going on. There are people at a crosswalk, in a city, and the weather is probably at a comfortable temperature. The system below recognized a few men, a few women, and jeans. The ability that people have to recognize what is going on in infinite numbers of'pictures' in our day-to-day lives is an ability formed by evolution, that scientists are trying to imitate through computer vision. The purpose of computer vision is to teach systems to recognize patterns in images so that they can process them in a similar way that people do.

NVIDIA's virtual GPU (vGPU) technology, which has already transformed virtual client computing, now supports server virtualization for AI, deep learning and data science. Previously limited to CPU-only, AI workloads can now be easily deployed on virtualized environments like VMware vSphere with new vComputeServer software and NVIDIA NGC. Through our partnership with VMware, this architecture will help organizations to seamlessly migrate AI workloads on GPUs between customer data centers and VMware Cloud on AWS. IT administrators can use hypervisor virtualization tools like VMware vSphere, including vCenter and vMotion, to manage all their data center applications, including AI applications running on NVIDIA GPUs. These GPU servers are often isolated, with the need to be managed separately.

Not every regression or classification problem needs to be solved with deep learning. For that matter, not every regression or classification problem needs to be solved with machine learning. After all, many data sets can be modeled analytically or with simple statistical procedures. On the other hand, there are cases where deep learning or deep transfer learning can help you train a model that is more accurate than you could create any other way. For these cases, PyTorch and TensorFlow can be quite effective, especially if there is already a trained model similar to what you need in the framework's model library.

Code that accompanies this article can be downloaded here. Over the years it became one of the most loved ML frameworks and gathered a massive amount of followers. Google has done a great job and incorporated this framework into Java, C, JavaScript and most importantly into major data science language Python. If you ask the community what are their favorite combination of tools, the most usual answer would be TensorFlow and Python. Thanks to this, we came to the point where this technology is mature enough to ease up its use and "cross the chasm". That is certainly one of the main goals of a new version – TensorFlow 2.0. However, let's start from the beginning and find out what is this technology all about.

The world famous Deep Learning Summit is returning to London for the 5th year, this time joined by the AI Assistant Summit and the AI in Retail and Advertising Summit. One pass provides access to all three events, plus the Deep Dive track. Confirmed speakers from Google, ASOS, DeepMind, Facebook AI Research, NASA, Amazon and more will come together to share the most cutting edge work in their space, bridging the gap between the latest technological research advancements and real world applications in business and society. Across the two day event, 90 speakers and 600 attendees will explore topics such as neural networks, machine learning, deep learning algorithms, speech recognition, robotics, personalised recommendation systems, image retrieval, reinforcement learning and more. Presentations from across the board touch on business case studies and research progressions, and the majority of attendees will spend time in all session rooms to expand their knowledge and find out how to apply AI in novel ways.