How can we understand progress in Deep Learning without a map? I created one such map a couple years ago, but this map needs a drastic overhaul. In "Five Capability Levels of Deep Learning Intelligence", I proposed a hierarchy of capabilities that was meant to inform the progress of Deep Learning development. So specifically, you begin with a feed forward network in the first level. That would be followed by memory enhanced networks, examples of which would include LSTM and Neural Turing Machine (NTM).

We've created an AI system, OpenAI Five, which has started to defeat amateur human teams. This video contains an overview of our system, some example gameplay, and professional caster Blitz's analysis of our bot, as we start to gear up for playing a professional team at this year's Dota world championships, The International. See our blog post for more details: https://blog.openai.com/openai-five.

Researchers have shown that they can use artificial intelligence (AI) to restore low-quality photos by exposing a neural network to only other low-quality photos, according to work presented at the International Conference on Machine Learning in Stockholm. The research was developed by representatives from Nvidia, a Santa Clara, California-based technology company, the Massachusetts Institute of Technology in Cambridge, Massachusetts, and Aalto University in Greater Helsinki, Finland. The team was able to clean up "grainy" photos by using deep learning to train the neural network with more than 50,000 example images, as explained in a news release from Nvidia. As the authors explained, their work has potential to be used in numerous industries, including radiology. "There are several real-world situations where obtaining clean training data is difficult: low-light photography (e.g., astronomical imaging), physically-based rendering, and magnetic resonance imaging," wrote author Jaakko Lehtinen, an associate professor at Aalto University, and colleagues.

Interest in artificial intelligence (AI) research has grown rapidly over the past few years, in part thanks to the numerous successes of modern machine learning techniques such as deep learning, the availability of large datasets and improvements in computing power. AI is proving to be increasingly applicable to healthcare and there is a growing list of tasks where algorithms have matched or surpassed physician performance. Despite the successes there remain significant concerns and challenges surrounding algorithm opacity, trust and patient data security. Notwithstanding these challenges, AI technologies will likely become increasingly integrated into emergency medicine in the coming years. This perspective presents an overview of current AI research relevant to emergency medicine.

Pretrained models are a wonderful source of help for people looking to learn an algorithm or try out an existing framework. Due to time restrictions or computational restraints, it's not always possible to build a model from scratch which is why pretrained models exist! You can use a pretrained model as a benchmark to either improve the existing model, or test your own model against it. The potential and possibilities are vast. In this article, we will look at various pretrained models in Keras that have applications in computer vision.

Artificial Intelligence (AI), a theory where machines perform tasks with intelligence like humans, has been the talk of the town across all industries and for all the right reasons. AI is no longer just used to describe the sophisticated consumer profiling platforms and techniques used by the likes of Google and Facebook, Amazon, etc. There has been widespread investigation on AI applications, research and now live adoption in fields like Healthcare, Manufacturing or Industrial applications, Logistics, Defense & Security, Retail, and many more in order to stay competitive while also taking advantage of the additional business models to generate additional revenue. However, the challenge lies in understanding what AI means in a business context and creating a converged architecture for the specific use case that is economically viable. At Fujitsu, we observed on our digital co-creation journey with several customers that IoT implementation in industrial enterprise applications is enabling customers to utilize AI to effectively automate their processes.

Autonomous cars are racing down the highway at speeds exceeding 100 MPH when suddenly a car a half-mile ahead blows out a tire sending dangerous debris across 3 lanes of traffic. Instead of relying upon sending this urgent, time-critical distress information to the world via the cloud, the cars on that particular section of the highway use peer-to-peer, immutable communications to inform all vehicles in the area of the danger so that they can slow down and move to unobstructed lanes (while also sending a message to the nearest highway maintenance robots to remove the debris). Real-time analytics at the edges of the Internet of Things and the real-time communications between devices of different types, models and makes are going to be critical to realizing the operational and society benefits of smart cities, smart airports, smart hospitals, smart factories and the like. While machine learning and reinforcement learning at the edges and deep learning at the core are key to driving system-wide intelligence, the ability to capture, communicate, share and build upon the system-wide "learnings and insights" at the "local level" is going to require a technology that exploits real-time, nearest neighbor, peer-to-peer, immutable communications. Hello Blockchain (see Figure 2)!

Four ways to explore the use of voice technology for your business. As Microsoft uses AI to chase down Amazon in retail, Google is drawing on its freakishly human-sounding Duplex technology to beat Microsoft in the call center. Google is hatching a plan to win over contact centers using its AI, announcing a slew of new features for its Dialogflow Enterprise Edition, which borrows from its own deep learning and AI research, as well as groundbreaking AI from Alphabet-owned DeepMind. The service, which launched in beta in April, aims to help customers build natural-sounding chatbots for the contact center, as well as phone-based conversational agents. It comes just days after Microsoft revealed its roadmap for Dynamics 365, including the new Dynamics 365 AI for Sales app, which will be in public preview in October, and is meant to help sales teams make decisions using call sentiment analysis and to deliver warnings about deals at risk.

It is now much easier than ever before to produce videos. While the ubiquitous video data is a great source for information discovery and extraction, the computational challenges are unparalleled. Automatically summarizing the videos has become a substantial need for browsing, searching, and indexing visual content. This paper is in the vein of supervised video summarization using sequential determinantal point process (SeqDPP), which models diversity by a probabilistic distribution. We improve this model in two folds. In terms of learning, we propose a large-margin algorithm to address the exposure bias problem in SeqDPP. In terms of modeling, we design a new probabilistic distribution such that, when it is integrated into SeqDPP, the resulting model accepts user input about the expected length of the summary. Moreover, we also significantly extend a popular video summarization dataset by 1) more egocentric videos, 2) dense user annotations, and 3) a refined evaluation scheme. We conduct extensive experiments on this dataset (about 60 hours of videos in total) and compare our approach to several competitive baselines.

One fundamental issue in managing bike sharing systems is the bike flow prediction. Due to the hardness of predicting the flow for a single station, recent research works often predict the bike flow at cluster-level. While such studies gain satisfactory prediction accuracy, they cannot directly guide some fine-grained bike sharing system management issues at station-level. In this paper, we revisit the problem of the station-level bike flow prediction, aiming to boost the prediction accuracy leveraging the breakthroughs of deep learning techniques. We propose a new multi-graph convolutional neural network model to predict the bike flow at station-level, where the key novelty is viewing the bike sharing system from the graph perspective. More specifically, we construct multiple inter-station graphs for a bike sharing system. In each graph, nodes are stations, and edges are a certain type of relations between stations. Then, multiple graphs are constructed to reflect heterogeneous relationships (e.g., distance, ride record correlation). Afterward, we fuse the multiple graphs and then apply the convolutional layers on the fused graph to predict station-level future bike flow. In addition to the estimated bike flow value, our model also gives the prediction confidence interval so as to help the bike sharing system managers make decisions. Using New York City and Chicago bike sharing data for experiments, our model can outperform state-of-the-art station-level prediction models by reducing 25.1% and 17.0% of prediction error in New York City and Chicago, respectively.