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]. If there are any problems/mistakes in these materials and slides, or there are something important/interesting you consider that should be added, just feel free to contact me.

My daughter has asked for a Windows laptop for Christmas, for schoolwork and games. I'm a Mac user and haven't set up a Windows machine for many years so I'd appreciate any advice ... except "get her a Mac/Linux" from below the line! Windows has changed a lot in the past decade, and now it's a mobile operating system. If your daughter has some experience with Google Android, she'll probably cope quite well. Many of Windows 10's main features came from the mobile world.

A few days ago I found out that there had appeared lda2vec (by Chris Moody) – a hybrid algorithm combining best ideas from well-known LDA (Latent Dirichlet Allocation) topic modeling algorithm and from a bit less well-known tool for language modeling named word2vec. And now I'm going to tell you a tale about lda2vec and my attempts to try it and compare with simple LDA implementation (I used gensim package for this). What is cool about it? It means that LDA is able to create document (and topic) representations that are not so flexible but mostly interpretable to humans. Also, LDA treats a set of documents as a set of documents, whereas word2vec works with a set of documents as with a very long text string.

When my mother was dying 22 years ago, the hospital briefly lost track of her location. Its error cost my family 40 precious minutes that we could have spent with her at the end of her life. The culprit in this horrific situation was the hospital's own hierarchy of processes and procedures that its workforce struggled to navigate during this emergency. I've always been passionate about the need for simplicity in the workplace. Simplicity lies at the foundation of organizations' empathy--policies that prioritize customers' and employees' experiences over its own interests--and that plays a key role in the future of work. When it comes to our attitudes and beliefs about the value of other people's time, particularly our employees' time on the job, we're still in the Industrial Era.

The latest artificial-intelligence techniques are being adopted by companies at a blistering pace. Before long, hackers might start taking a closer look, too, and they could cause all sorts of trouble by tricking these systems with illusory data. Speaking at a recent AI conference in Barcelona, Spain, Ian Goodfellow, a research scientist at OpenAI who has done pioneering work on deceiving machine-learning systems, said attacking the systems is easy. "Almost anything bad you can think of doing to a machine-learning model can be done right now," he said. "And defending is really, really hard."

Rob May is the CEO and Co-Founder of Talla, a platform for intelligent information delivery in Slack and Hipchat. Previously, Rob was the CEO and Co-Founder of Backupify, (acquired by Datto in 2014). Before that, he held engineering, business development, and management positions at various startups. Rob has a B.S. in Electrical Engineering and a MBA from the University of Kentucky. He is also a well known angel investor, a venture partner at Pillar, and is the creator and writer of the widely-read and highly-regarded AI newsletter, Technically Sentient. Rob May will be speaking at the Gigaom AI Now in San Francisco, February 15-16th. In anticipation of that, I caught up with him to ask a few questions.

Only a few short years ago, healthcare organizations were wondering what exactly "big data" was and why they had to care about it. As the industry moves into 2017, they might have similar questions about the definitions of terms like "blockchain," "the Internet of Things," and "artificial intelligence" – but the use cases for these cutting-edge technologies are rapidly becoming crystal clear. From precision medicine and business intelligence to data security and patient engagement, the IoT, AI, and blockchain hold exciting promises for providers, patients, and researchers looking to move their big data hoards from repositories to real results. The healthcare sector must join its peers in other industries to leverage these new applications for big data, according to a pair of reports fromthe White House and Gartner, Inc., in order to take advantage of the nearly-limitless opportunities for lowering costs, improving outcomes, and achieving quality goals. Artificial intelligence will soon "improve the world" While android physicians and self-driving gurneys are likely still several decades away, basic artificial intelligence programs are already making an impact on everyday society.

The term "artificial intelligence" historically refers to systems that attempt to mimic or replicate human thought. This is not an accurate description of the actual science of artificial intelligence, and it implies a false choice between artificial and natural intelligences. That is why IBM and others have chosen to use different language to describe our work in this field. We feel that "cognitive computing" or "augmented intelligence" -- which describes systems designed to augment human thought, not replicate it -- are more representative of our approach. There is little commercial or societal imperative for creating "artificial intelligence."

He's a lightning-fast learner, he speaks eight languages and he's considered an expert in multiple fields. He's got an exemplary work ethic, is a speed reader and finds insights no one else can. On a personal note, he's a mean chef and even offers good dating advice. Named after IBM's first CEO, Watson was born back in 2007 as part of an effort by IBM Research to develop a question-answering system that could compete on the American quiz show "Jeopardy." Since trouncing its human opponents on the show in 2011, it has expanded considerably.

Whether you're an avid Starbucks, coffee, or just a tech fan, this new upgraded Starbucks App will make your day. As one of the most loved coffee providers, there is, Starbucks just found another new and funky way to keep their customers both entertained and satisfied, and that's by adding a barista chatbot to their app. It's simply another excellent example of the good things that AI has to offer and how it will make our lives that little bit better. This new feature is aptly named My Starbucks Barista and is "an innovative conversational ordering system." There will be a limited beta version of the feature that's expected to be released in early 2017 that will enable customers to place their orders via a chatbot interface or simple voice command.