We give a novel formal theoretical framework for unsupervised learning with two distinctive characteristics. First, it does not assume any generative model and based on a worst-case performance metric. Second, it is comparative, namely performance is measured with respect to a given hypothesis class. This allows to avoid known computational hardness results and improper algorithms based on convex relaxations. We show how several families of unsupervised learning models, which were previously only analyzed under probabilistic assumptions and are otherwise provably intractable, can be efficiently learned in our framework by convex optimization.

In this paper, we prove a conjecture published in 1989 and also partially address an open problem announced at the Conference on Learning Theory (COLT) 2015. With no unrealistic assumption, we first prove the following statements for the squared loss function of deep linear neural networks with any depth and any widths: 1) the function is non-convex and non-concave, 2) every local minimum is a global minimum, 3) every critical point that is not a global minimum is a saddle point, and 4) there exist "bad" saddle points (where the Hessian has no negative eigenvalue) for the deeper networks (with more than three layers), whereas there is no bad saddle point for the shallow networks (with three layers). Moreover, for deep nonlinear neural networks, we prove the same four statements via a reduction to a deep linear model under the independence assumption adopted from recent work. As a result, we present an instance, for which we can answer the following question: how difficult is it to directly train a deep model in theory? It is more difficult than the classical machine learning models (because of the non-convexity), but not too difficult (because of the nonexistence of poor local minima). Furthermore, the mathematically proven existence of bad saddle points for deeper models would suggest a possible open problem. We note that even though we have advanced the theoretical foundations of deep learning and non-convex optimization, there is still a gap between theory and practice.

While many models are purposed for detecting the occurrence of significant events in financial systems, the task of providing qualitative detail on the developments is not usually as well automated. We present a deep learning approach for detecting relevant discussion in text and extracting natural language descriptions of events. Supervised by only a small set of event information, comprising entity names and dates, the model is leveraged by unsupervised learning of semantic vector representations on extensive text data. We demonstrate applicability to the study of financial risk based on news (6.6M articles), particularly bank distress and government interventions (243 events), where indices can signal the level of bank-stress-related reporting at the entity level, or aggregated at national or European level, while being coupled with explanations. Thus, we exemplify how text, as timely, widely available and descriptive data, can serve as a useful complementary source of information for financial and systemic risk analytics.

In second place, Ng saw neither unsupervised learning nor reinforcement learning, but transfer learning. One of the hottest developments within Deep Learning was Generative Adversarial Networks (GANs). Secondly, end-to-end (supervised) Deep Learning allows us to learn to map from inputs directly to outputs. The Conference on Neural Information Processing Systems (NIPS) is one of the two top conferences in machine learning. Among ML research areas, supervised learning is the undisputed driver of the recent success of ML and will likely continue to drive it for the foreseeable future.

We alluded to the possibility of Deep Learning and IoT previously where we said that Deep learning algorithms play an important role in IoT analytics because Machine data is sparse and / or has a temporal element to it. Devices may behave differently at different conditions. Hence, capturing all scenarios for data pre-processing/training stage of an algorithm is difficult. Deep learning algorithms can help to mitigate these risks by enabling algorithms learn on their own. This concept of machines learning on their own can be extended to machines teaching other machines.

To learn more about the state of AI today and where we might be headed in coming years, download the free report "What is Artificial Intelligence?," by Mike Loukides and Ben Lorica. There are four basic ingredients for making AI: data, compute resources (i.e., hardware), algorithms (i.e., software), and the talent to put it all together. In this era of deep learning ascendancy, it has become conventional wisdom that data is the most differentiating and defensible of these resources; companies like Google and Facebook spend billions to develop and provide consumer services, largely in order to amass information about their users and the world they inhabit. While the original strategic motivation behind these services was to monetize that data via ad targeting, both of these companies--and others who are desperate to follow their lead--now view the creation of AI as an equally important justification for their massive collection efforts. While all four pieces are necessary to build modern AI systems, what we'll call their "scarcity" varies widely.

It is something that companies and businesses have been trying to implement (and something that society has feared) for decades. However, with all the recent advancements to democratize artificial intelligence and use it for good, almost every company started to turn to this technology and technique in 2016. The year started with Facebook's CEO Mark Zuckerberg announcing his plan to build an artificially intelligent assistant to do everything from adjusting the temperature in his house to checking up on his baby girl. He worked throughout the year to bring his plan to life, with an update in August that stated he was almost ready to show off his AI to the world. In November, Facebook announced it was beginning to focus on giving computers the ability to think, learn, plan and reason like humans.

With the advent of drones, robots and self-driving vehicles a recent report released by the White House contended that artificial intelligence could transform the economy. The report lists strategies that will increase the benefits and diminish the costs of AI, helping to create opportunities in cyberdefense and fraud detection. Although the report clearly states that the precise economic impact is difficult to estimate, it highlights five possible effects of an economy driven by AI, including an increased demand for higher technical skills and uneven distribution of the impact across sectors, wage levels, etc. A depiction of AI has often been seen on television shows. In 1962, I sing the Body Electric, an episode of The Twilight Zone, significantly captured the unique concept of AI by presenting a robotic grandmother adjusting smoothly in a human family.

Scientists at the Pharmaceutical Artificial Intelligence (pharma.AI) group of Insilico Medicine, Inc, have announced the publication of a seminal paper demonstrating the application of generative adversarial autoencoders (AAEs) to generating new molecular fingerprints on demand. The study was published in Oncotarget on 22nd of December, 2016. The study represents the proof of concept for applying Generative Adversarial Networks (GANs) to drug discovery. The authors significantly extended this model to generate new leads according to multiple requested characteristics and plan to launch a comprehensive GAN-based drug discovery engine producing promising therapeutic treatments to significantly accelerate pharmaceutical R&D and improve the success rates in clinical trials. Since 2010 deep learning systems demonstrated unprecedented results in image, voice and text recognition, in many cases surpassing human accuracy and enabling autonomous driving, automated creation of pleasant art and even composition of pleasant music.

Fast food restaurant chain Kentucky Fried Chicken (KFC) launched its first Artificial Intelligence (AI)-enabled cafe in Beijing, and said it plans to create more innovative and interesting dining experiences for customers. With the cooperation of Baidu Inc, China's largest search engine, KFC started its first smart restaurant in the Financial Street area in Beijing on Friday, the People's Daily reported on Sunday. At the cafe, customers are able to take pictures with a machine, which will recognise the diner's face, sex, age, mood and other features, then help to recommend suitable food and set meals and complete the ordering process. "If the consumer visits the store again and takes a picture with the machine, it will be able to recognize his or her face and show the previous purchase history, remember the customer's dining habits, and help to place an order faster," said Wu Zhongqin, Deputy Director of the Institute of Deep Learning of Baidu Inc, which helped to develop the technology. With another machine with an augmented reality, customers are able to interact with the machine, change facial expressions by shaking their heads in front of the machine, take photos, and save them to their phones.