Genome-wide gene deletion studies have shown that a small fraction of genes in a genome are indispensable to the survival or reproduction of an organism (Winzeler et al., 1999; Kamath et al., 2003). These genes are referred as essential genes, and essential proteins are the products of essential genes. The deletion of such essential genes will result in lethality or infertility. Since being essential is highly dependent on the circumstances in which an organism lives, recently systematic attempts have been made to identify those genes that are absolutely required to maintain life, provided that all nutrients are available (Zhang and Lin, 2009). Such experiments have led to the conclusion that the absolutely required number of genes for a bacteria is on the order of about 250–300, which encode proteins to maintain a central metabolism, replicate DNA, translate genes into proteins, maintain a basic cellular structure, and mediate transport processes into and out of the cell.

For nearly 30 years, Pedro Domingos has been working within artificial intelligence communities, both as a researcher and a developer. In that time, the University of Washington professor has become an expert on all things machine learning. His new book, The Master Algorithm, traces the beginnings of machine learning, and provides a road-map to where it may be headed in the future. We talked to him by phone about what lies ahead for machine learning, how the master algorithm might change everything, and why he thinks the idea of the singularlity is overstated. Can you give me the elevator pitch on what "The Master Algorithm" is?

In a digital classroom, analysis of students' interactions with the learning media provides important information about users' behavior, which can lead to a better understanding and thus optimizes teaching and learning. However, over the period of a course, students tend to forget the lessons learned in class. Learning predictions can be used to recommend learning objects users need most, as well as to give an overview of current knowledge and the learning level. The representation of time based data in such a format is difficult since the knowledge level of a user with a learning object changes continuously depending on various factors. This paper presents work in progress for a doctoral approach to extend the traditional user-item-matrix of a recommendation engine by a third dimension - the time value. Moreover, in this approach the learning need consists of different context factors each influencing the relevance score of a learning object.

Guest blog post by Xiu-Shen Wei, originally posted here. Deep Neural Networks, especially Convolutional Neural Networks (CNN), allows computational models that are composed of multiple processing layers to learn representations of data with multiple levels of abstraction. These methods have dramatically improved the state-of-the-arts in visual object recognition, object detection, text recognition and many other domains such as drug discovery and genomics. In addition, many solid papers have been published in this topic, and some high quality open source CNN software packages have been made available. There are also well-written CNN tutorials or CNN software manuals. However, it might lack a recent and comprehensive summary about the details of how to implement an excellent deep convolutional neural networks from scratch. Thus, we collected and concluded many implementation details for DCNNs. Here we will introduce these extensive implementation details, i.e., tricks or tips, for building and training your own deep networks. We assume you already know the basic knowledge of deep learning, and here we will present the implementation details (tricks or tips) in Deep Neural Networks, especially CNN for image-related tasks, mainly in eight aspects: 1) data augmentation; 2) pre-processing on images; 3) initializations of Networks; 4) some tips during training; 5) selections of activation functions; 6) diverse regularizations; 7)some insights found from figures and finally 8) methods of ensemble multiple deep networks. Additionally, the corresponding slides are available at [slide].

Network analysis offers a new set of techniques to tackle the persistent and growing problem of complex fraud. Network analysis supplements traditional techniques by providing a mechanism to bridge investigative and analytics methods. Beyond base visualization, network analysis provides a standardized platform for complex fraud pattern storage and retrieval, pattern discovery and detection, statistical analysis, and risk scoring. This article gives an overview of the main challenges and demonstrates a promising approach using a hands-on example. With swelling globalization, advanced digital communication technology, and international financial deregulation, fraud investigators face a daunting battle against increasingly sophisticated fraudsters.

The stage and grade of psoriasis severity is clinically relevant and important for dermatologists as it aids them lead to a reliable and an accurate decision making process for better therapy. This paper proposes a novel psoriasis risk assessment system (pRAS) for stratification of psoriasis severity from colored psoriasis skin images having Asian Indian ethnicity. Machine learning paradigm is adapted for risk stratification of psoriasis disease grades utilizing offline training and online testing images. It uses two kinds of classifiers (support vector machines (SVM) and decision tree (DT)) during training and testing phases and two kinds of feature selection criteria (Principal Component Analysis (PCA) and Fisher Discriminant Ratio (FDR)), thus, leading to an exhaustive comparison between these four systems. Our database consisted of 848 psoriasis images with five severity grades: healthy, mild, moderate, severe and very severe, consisting of 383, 47, 245, 145, and 28 images respectively.

I recently gave a talk about recommender systems at the Data Science Sydney meetup (the slides are available here). This post roughly follows the outline of the talk, expanding on some of the key points in non-slide form (i.e., complete sentences and paragraphs!). The first few sections give a broad overview of the field and the common recommendation paradigms, while the final part is dedicated to debunking five common myths about recommender systems. The key reason why many people seem to care about recommender systems is money. For companies such as Amazon, Netflix, and Spotify, recommender systems drive significant engagement and revenue. But this is the more cynical view of things.

Alphabet founders Larry Page and Sergey Brin are so enamored with the guy they put in charge of Google, CEO Sundar Pichai, that they turned him into an honorary founder and let him write their annual Founder's Letter this year. The Founder's Letter is traditionally when the founders spell out their vision for the company. But Page explained that he's so "pleased with Sundar's performance" and because "the majority of our big bets are in Google" that he decided to give Pichai "the bully-pulpit here" to talk about Google. A more cynical take would be that Page also wanted to sidestep talking about the controversies plaguing the other side of Alphabet, the so-called "Other Bets" division outside of the Google unit. As Business Insider's Jillian D'Onfro reports, although Alphabet makes 99% of its revenue from Google, mostly from its gargantuan ads businesses, it has been the "Other Bets" that have attracted the most media attention in recent months.

Speaking after the first day of the ICT meeting, Takaichi said she introduced eight basic principles Tokyo believes important when developing computer science that gives machines human-like intelligence, and that she was generally supported in calling for further discussion. The eight principles include making AI networks controllable by human beings and respect for human dignity and privacy. "The development of AI is expected to progress at a tremendous pace of speed, and it should be amazing technology that does not give anxiety to people," the minister of internal affairs and communications told reporters, noting the need to deepen international discussion about establishing a basic set of rules. The first G-7 ICT ministerial meeting in nearly two decades comes at a time when cyberattacks have become a global reality and the development of such potentially revolutionary technologies as artificial intelligence and the "Internet of Things" (IoT) -- the concept of connecting various products to the Internet -- continues apace. With cyberattacks having become a global reality, participants from Britain, Canada, France, Germany, Italy, Japan and the United States discussed at the G-7 meeting ways to utilize advances in the field to drive economic growth while ensuring data security.

Using humanoid robots to study human behavior.