While perception tasks such as visual object recognition and text understanding play an important role in human intelligence, the subsequent tasks that involve inference, reasoning and planning require an even higher level of intelligence. The past few years have seen major advances in many perception tasks using deep learning models. For higher-level inference, however, probabilistic graphical models with their Bayesian nature are still more powerful and flexible. To achieve integrated intelligence that involves both perception and inference, it is naturally desirable to tightly integrate deep learning and Bayesian models within a principled probabilistic framework, which we call Bayesian deep learning. In this unified framework, the perception of text or images using deep learning can boost the performance of higher-level inference and in return, the feedback from the inference process is able to enhance the perception of text or images. This paper proposes a general framework for Bayesian deep learning and reviews its recent applications on recommender systems, topic models, and control. In this paper, we also discuss the relationship and differences between Bayesian deep learning and other related topics like Bayesian treatment of neural networks.

This paper describes a series of new qualitative spatial logics for checking consistency of sameAs and partOf matches between spatial objects from different geospatial datasets, especially from crowd-sourced datasets. Since geometries in crowd-sourced data are usually not very accurate or precise, we buffer geometries by a margin of error or a level of tolerance, and define spatial relations for buffered geometries. The spatial logics formalize the notions of `buffered equal' (intuitively corresponding to `possibly sameAs'), `buffered part of' (`possibly partOf'), `near' (`possibly connected') and `far' (`definitely disconnected'). A sound and complete axiomatisation of each logic is provided with respect to models based on metric spaces. For each of the logics, the satisfiability problem is shown to be NP-complete. Finally, we briefly describe how the logics are used in a system for generating and debugging matches between spatial objects, and report positive experimental evaluation results for the system.

The author is a Silicon Valley based IIoT industry analyst & co-founder of ArcInsight Research Partners, a research & advisory group. He has also trained with business strategy consulting firms, with additional qualifications in decision analytics, Bayesian-learning approaches & risk-assessment. I wrote about GE's Predix Platform last year in the context of its immensely successful Minds Machines Conferences where the company showcased a very compelling story about its quiet but steady transformation into a digital industrial company. This is a far cry from creating avatars in a consumer or mobile context. Industrial stakes are very high.

Sum-Product Networks (SPNs) are recently introduced deep tractable probabilistic models by which several kinds of inference queries can be answered exactly and in a tractable time. Up to now, they have been largely used as black box density estimators, assessed only by comparing their likelihood scores only. In this paper we explore and exploit the inner representations learned by SPNs. We do this with a threefold aim: first we want to get a better understanding of the inner workings of SPNs; secondly, we seek additional ways to evaluate one SPN model and compare it against other probabilistic models, providing diagnostic tools to practitioners; lastly, we want to empirically evaluate how good and meaningful the extracted representations are, as in a classic Representation Learning framework. In order to do so we revise their interpretation as deep neural networks and we propose to exploit several visualization techniques on their node activations and network outputs under different types of inference queries. To investigate these models as feature extractors, we plug some SPNs, learned in a greedy unsupervised fashion on image datasets, in supervised classification learning tasks. We extract several embedding types from node activations by filtering nodes by their type, by their associated feature abstraction level and by their scope. In a thorough empirical comparison we prove them to be competitive against those generated from popular feature extractors as Restricted Boltzmann Machines. Finally, we investigate embeddings generated from random probabilistic marginal queries as means to compare other tractable probabilistic models on a common ground, extending our experiments to Mixtures of Trees.

With the explosion of data generation, getting optimal solutions to data driven problems is increasingly becoming a challenge, if not impossible.The importance of machine learning algorithms, which can handle this burst of data and assist in intelligent decision making, is thus realised among data scientists. Within this category of machine learning algorithms, a special focus area is bio-inspired algorithms. This review article provides the readers some inputs on the advances in the domain of bio inspired algorithms and their potential applications across domains. It is increasingly being recognised that applications of intelligent bio-inspired algorithms are necessary for addressing highly complex problems to provide working solutions in time, especially with dynamic problem definitions, fluctuations in constraints, incomplete or imperfect information and limited computation capacity. More and more such intelligent algorithms are thus being explored for solving different complex problems. While some studies are exploring the application of these algorithms in a novel context, other studies are incrementally improving the algorithm itself.

Developing intelligent systems requires combining results from both industry and academia. In this report you find an overview of relevant research fields and industrially applicable technologies for building very large scale cyber physical systems. A concept architecture is used to illustrate how existing pieces may fit together, and the maturity of the subsystems is estimated. The goal is to structure the developments and the challenge of machine intelligence for Consumer and Industrial Internet technologists, cyber physical systems researchers and people interested in the convergence of data & Internet of Things. It can be used for planning developments of intelligent systems.

The future of Education is always a hot topic anywhere in the globe, as we universally want the best for the next generation. As I get older, I see a rapidly growing gap between how I learnt at school and at University, and what the current working world expects. Recently, I was on a speaking panel for a conference about Emerging Trends in Learning and working tackling the practical impacts of the digital disruption starting to hit the Education sector. Greg Prior from NSW Education, drove the point that literacy and numeracy are still a fundamental core. Collaboration and personalised learning is emerging as a key approach and governments will allow students to progressively take leadership of their own learning.

Artificial intelligence is proving itself to be more than just a buzzword recently, especially when it comes to marketing. From IBM Watson to the rise and rise of chatbots, AI's not only becoming prominent in largescale marketing channels everywhere, but it's making headway in video advertising too, it would seem. That's at least according to video platform LoopMe, which has announced new optimisation and reporting technology, PurchaseLoop. The new product combines LoopMe's AI with third party brand research from Nielsen and On Device, to target users and moments which have shown the greatest likelihood of moving along the purchase funnel. Throughout each campaign, a sample of exposed users will be surveyed and benchmarked against a control to determine the campaign's ongoing performance against brand metrics like brand awareness, favourability or purchase intent.

A global hub of the machinery industry, Taiwan is stepping up efforts to develop innovative smart manufacturing technologies. From May 20-24 this year, about 30,000 visitors, including buyers from Asia, Europe and the U.S., flocked to the Commercial Exhibition Center in Taichung City, central Taiwan for the Automatic Machinery and Intelligent Manufacturing Exhibition. The trade show has been staged annually in Taichung for more than three decades, though this marked the first time that intelligent manufacturing was used in the title of the event. Organizer Commercial Times, one of the country's two major financial newspapers, opted to alter the name of the show to highlight the growing focus on this field in the nation's globally competitive machinery sector. "This is the 32nd edition of our machinery show in Taichung. But unlike previous events, this year's exhibition features intelligent machines to reflect the current trend in manufacturing systems development," Chen Kuo-wei (???), president of Commercial Times, said at the opening ceremony of the five-day event, which generated business deals totaling about NT 300 million (US 9.2 million).

While the Matrix Generalized Inverse Gaussian ($\mathcal{MGIG}$) distribution arises naturally in some settings as a distribution over symmetric positive semi-definite matrices, certain key properties of the distribution and effective ways of sampling from the distribution have not been carefully studied. In this paper, we show that the $\mathcal{MGIG}$ is unimodal, and the mode can be obtained by solving an Algebraic Riccati Equation (ARE) equation [7]. Based on the property, we propose an importance sampling method for the $\mathcal{MGIG}$ where the mode of the proposal distribution matches that of the target. The proposed sampling method is more efficient than existing approaches [32, 33], which use proposal distributions that may have the mode far from the $\mathcal{MGIG}$'s mode. Further, we illustrate that the the posterior distribution in latent factor models, such as probabilistic matrix factorization (PMF) [25], when marginalized over one latent factor has the $\mathcal{MGIG}$ distribution. The characterization leads to a novel Collapsed Monte Carlo (CMC) inference algorithm for such latent factor models. We illustrate that CMC has a lower log loss or perplexity than MCMC, and needs fewer samples.