Declarative modeling uses symbolic expressions to represent models. With such expressions one can formalize high-level mathematical computations on models that would be difficult or impossible to perform directly on a lower-level simulation program, in a general-purpose programming language. Examples of such computations on models include model analysis, relatively general-purpose model-reduction maps, and the initial phases of model implementation, all of which should preserve or approximate the mathematical semantics of a complex biological model. Multiscale modeling benefits from both the expressive power of declarative modeling languages and the application of model reduction methods to link models across scale. Based on previous work, here we define declarative modeling of complex biological systems by defining the semantics of an increasingly powerful series of declarative modeling languages including reaction-like dynamics of parameterized and extended objects, we define semantics-preserving implementation and semantics-approximating model reduction transformations, and we outline a "meta-hierarchy" for organizing declarative models and the mathematical methods that can fruitfully manipulate them.

Every year, more than a billion people around the world celebrate Chinese New Year and engage in a subtle linguistic dance with luck. You can think of it as a set of holiday rituals that resemble a courtship. To lure good fortune into their lives, they may decorate their homes and doors with paper cutouts of lucky words or phrases. Those who need a haircut make sure to get one before the New Year, as the word for "hair" (fa) sounds like the word for "prosperity"--and who wants to snip away prosperity, even if it's just a trim? The menu of food served at festive meals often includes fish, because its name (yu) sounds the same as the word for "surplus"; a type of algae known as fat choy because in Cantonese it sounds like "get rich"; and oranges, because in certain regions their name sounds like the word for "luck."

I have two children, and they are a study in contrasts: My son works at a gym designing and building rock-climbing walls; In his spare time, he climbs them. My daughter is a Ph.D. student in immunology; In her spare time, she writes novels. My son is the sort of person you want around in a crisis, cool-headed and springing to action. Let's just say my daughter is not. My son spends money as soon as he earns it.

The machine reading task, where a computer reads a document and answers questions about it, is important in artificial intelligence research. Recently, many models have been proposed to address it. Word-level models, which have words as units of input and output, have proven to yield state-of-the-art results when evaluated on English datasets. However, in morphologically richer languages, many more unique words exist than in English due to highly productive prefix and suffix mechanisms. This may set back word-level models, since vocabulary sizes too big to allow for efficient computing may have to be employed. Multiple alternative input granularities have been proposed to avoid large input vocabularies, such as morphemes, character n-grams, and bytes. Bytes are advantageous as they provide a universal encoding format across languages, and allow for a small vocabulary size, which, moreover, is identical for every input language. In this work, we investigate whether bytes are suitable as input units across morphologically varied languages. To test this, we introduce two large-scale machine reading datasets in morphologically rich languages, Turkish and Russian. We implement 4 byte-level models, representing the major types of machine reading models and introduce a new seq2seq variant, called encoder-transformer-decoder. We show that, for all languages considered, there are models reading bytes outperforming the current state-of-the-art word-level baseline. Moreover, the newly introduced encoder-transformer-decoder performs best on the morphologically most involved dataset, Turkish. The large-scale Turkish and Russian machine reading datasets are released to public.

In his session at 21st Cloud Expo, James Henry, Co-CEO/CTO of Calgary Scientific Inc., introduced you to the challenges, solutions and benefits of training AI systems to solve visual problems with an emphasis on improving AIs with continuous training in the field. He explored applications in several industries and discuss technologies that allow the deployment of advanced visualization solutions to the cloud. Speaker Bio James Henry is Co-CEO/CTO of Calgary Scientific Inc., a company specializing in bringing real time interactive software to cloud and mobile platforms. He has 25 years of experience leading software teams in many industries including the oil and gas, healthcare, telecommunication, geolocation, construction and simulation industries. His current interest is in enabling people, data and AIs to interact in real time to solve complex problems.

This paper describes a novel text-to-speech (TTS) technique based on deep convolutional neural networks (CNN), without any recurrent units. Recurrent neural network (RNN) has been a standard technique to model sequential data recently, and this technique has been used in some cutting-edge neural TTS techniques. However, training RNN component often requires a very powerful computer, or very long time typically several days or weeks. Recent other studies, on the other hand, have shown that CNN-based sequence synthesis can be much faster than RNN-based techniques, because of high parallelizability. The objective of this paper is to show an alternative neural TTS system, based only on CNN, that can alleviate these economic costs of training. In our experiment, the proposed Deep Convolutional TTS can be sufficiently trained only in a night (15 hours), using an ordinary gaming PC equipped with two GPUs, while the quality of the synthesized speech was almost acceptable.

In his session at 21st Cloud Expo, James Henry, Co-CEO/CTO of Calgary Scientific Inc., will introduce you to the challenges, solutions and benefits of training AI systems to solve visual problems with an emphasis on improving AIs with continuous training in the field. He will explore applications in several industries and discuss technologies that allow the deployment of advanced visualization solutions to the cloud. Speaker Bio James Henry is Co-CEO/CTO of Calgary Scientific Inc., a company specializing in bringing real time interactive software to cloud and mobile platforms. He has 25 years of experience leading software teams in many industries including the oil and gas, healthcare, telecommunication, geolocation, construction and simulation industries. His current interest is in enabling people, data and AIs to interact in real time to solve complex problems.

Optimization algorithms for training deep models not only affects the convergence rate and stability of the training process, but are also highly related to the generalization performance of the models. While adaptive algorithms, such as Adam and RMSprop, have shown better optimization performance than stochastic gradient descent (SGD) in many scenarios, they often lead to worse generalization performance than SGD, when used for training deep neural networks (DNNs). In this work, we identify two problems of Adam that may degrade the generalization performance. As a solution, we propose the normalized direction-preserving Adam (ND-Adam) algorithm, which combines the best of both worlds, i.e., the good optimization performance of Adam, and the good generalization performance of SGD. In addition, we further improve the generalization performance in classification tasks, by using batch-normalized softmax. This study suggests the need for more precise control over the training process of DNNs.

DeepMind Alberta will turbo-charge the research ecosystem, mirroring the partnerships DeepMind has nurtured with the top academic institutions in the UK,

Just one look at your smartphone is all it takes to remind you that digitization and automation in financial services is nothing new. But the recent heightened interest in artificial intelligence (AI) and banking is. "The explosive growth of structured and unstructured data, availability of new technologies such as cloud computing and machine learning algorithms, rising pressures brought by new competition, increased regulation and heightened consumer expectations"--all of these factors, he says--"have created a'perfect storm' for the expanded use of artificial intelligence in financial services." For many financial industry leaders, understanding how AI can be incorporated into their business operations can be a storm in and of itself. With this in mind, we reached out to industry thought leaders in advance of October's BAI Beacon financial services conference, where AI is going to be a huge topic of conversation in a slate of Innovation & FinTech sessions, and asked: "What can banks and financial institutions (FIs) do right now to get started with AI in their business?"