Many problems in machine learning involve calculating correspondences between sets of objects, such as point clouds or images. Discrete optimal transport (OT) provides a natural and successful approach to such tasks whenever the two sets of objects can be represented in the same space or when we can evaluate distances between the objects. Unfortunately neither requirement is likely to hold when object representations are learned from data. Indeed, automatically derived representations such as word embeddings are typically fixed only up to some global transformations, for example, reflection or rotation. As a result, pairwise distances across the two types of objects are ill-defined without specifying their relative transformation. In this work, we propose a general framework for optimal transport in the presence of latent global transformations. We discuss algorithms for the specific case of orthonormal transformations, and show promising results in unsupervised word alignment.

The current trend of pushing CNNs deeper with convolutions has created a pressing demand to achieve higher compression gains on CNNs where convolutions dominate the computation and parameter amount (e.g., GoogLeNet, ResNet and Wide ResNet). Further, the high energy consumption of convolutions limits its deployment on mobile devices. To this end, we proposed a simple yet effective scheme for compressing convolutions though applying k-means clustering on the weights, compression is achieved through weightsharing, by only recording K cluster centers and weight assignment indexes. We then introduced a novel spectrally relaxed k-means regularization, which tends to make hard assignments of convolutional layer weights to K learned cluster centers during retraining. We additionally propose an improved set of metrics to estimate energy consumption of CNN hardware implementations, whose estimation results are verified to be consistent with previously proposed energy estimation tool extrapolated from actual hardware measurements. We finally evaluated Deep k-Means across several CNN models in terms of both compression ratio and energy consumption reduction, observing promising results without incurring accuracy loss. The code is available at https://github.

Machine learning is revolutionizing the way we interact with the world. Popular websites use algorithms to analyze user data and recommend videos, customize social media feeds, and optimize advertisements. Unsurprisingly, machine learning is taking a large role in making decisions about human beings, ranging from credit to parole decisions, and is likely to be more and more widely used in the future. It is not hard to imagine that, even in cases where the final decisions are made by people, they will be doing so with advice from algorithms that make inferences from patterns in petabytes of data. Some proponents of machine learning have suggested that not only are these algorithms able to leverage the increasing amount of data we have access to, but also that they might be able to make these decisions more fairly, as they seem to not be subject to human biases. There is some truth to these claims.

We show that Thompson Sampling with Gaussian Prior as detailed by Algorithm 2 in (Agrawal & Goyal, 2013) is already differentially private. Theorem 1 show that it enjoys a very competitive privacy loss of only $\mathcal{O}(\ln^2 T)$ after T rounds. Finally, Theorem 2 show that one can control the privacy loss to any desirable $\epsilon$ level by appropriately increasing the variance of the samples from the Gaussian posterior. And this increases the regret only by a term of $\mathcal{O}(\frac{\ln^2 T}{\epsilon})$. This compares favorably to the previous result for Thompson Sampling in the literature ((Mishra & Thakurta, 2015)) which adds a term of $\mathcal{O}(\frac{K \ln^3 T}{\epsilon^2})$ to the regret in order to achieve the same privacy level. Furthermore, our result use the basic Thompson Sampling with few modifications whereas the result of (Mishra & Thakurta, 2015) required sophisticated constructions.

We present here a new probabilistic inference algorithm that gives exact results in the domain of discrete probability distributions. This algorithm, named the Statues algorithm, calculates the marginal probability distribution on probabilistic models defined as direct acyclic graphs. These models are made up of well-defined primitives that allow to express, in particular, joint probability distributions, Bayesian networks, discrete Markov chains, conditioning and probabilistic arithmetic. The Statues algorithm relies on a variable binding mechanism based on the generator construct, a special form of coroutine; being related to the enumeration algorithm, this new algorithm brings important improvements in terms of efficiency, which makes it valuable in regard to other exact marginalization algorithms. After introduction of several definitions, primitives and compositional rules, we present in details the Statues algorithm. Then, we briefly discuss the interest of this algorithm compared to others and we present possible extensions. Finally, we introduce Lea and MicroLea, two Python libraries implementing the Statues algorithm, along with several use cases.

Connected vehicle fleets are deployed worldwide in several industrial IoT scenarios. With the gradual increase of machines being controlled and managed through networked smart devices, the predictive maintenance potential grows rapidly. Predictive maintenance has the potential of optimizing uptime as well as performance such that time and labor associated with inspections and preventive maintenance are reduced. In order to understand the trends of vehicle faults with respect to important vehicle attributes viz mileage, age, vehicle type etc this problem is addressed through hierarchical modified fuzzy support vector machine (HMFSVM). The proposed method is compared with other commonly used approaches like logistic regression, random forests and support vector machines. This helps better implementation of telematics data to ensure preventative management as part of the desired solution. The superiority of the proposed method is highlighted through several experimental results.

We introduce a principled approach for unsupervised structure learning of deep neural networks. We propose a new interpretation for depth and inter-layer connectivity where conditional independencies in the input distribution are encoded hierarchically in the network structure. Thus, the depth of the network is determined inherently (equal to the maximal order of independence in the input distribution). The proposed method casts the problem of neural network structure learning as a problem of Bayesian network structure learning. Then, instead of directly learning the discriminative structure, it learns a generative graph, constructs its stochastic inverse, and then constructs a discriminative graph. We prove that conditional-dependency relations among the latent variables in the generative graph are preserved in the class-conditional discriminative graph. We demonstrate on image classification benchmarks that the deepest layers (convolutional and dense) of common networks can be replaced by significantly smaller learned structures, while maintaining classification accuracy---state-of-the-art on tested benchmarks. Our structure learning algorithm requires a small computational cost and runs efficiently on a standard desktop CPU.

"We have the opportunity in this election in Brazil for the first time, here and around the world, to be very prepared to deal with the pitfalls of technology, such as fake news, social bots and macro-targets," said Rodrigo Helcer, CEO of Stilingue, a technology company specialized in artificial intelligence, during the talk "AI and Elections in Brazil" at the Path Festival in S--o Paulo. Stilingue was created to monitor social media posts and the media in Portuguese using artificial intelligence (AI). During the elections, marketing and advertising companies will use Stilingue technology to promote candidates and to help manage politicians' reputations. "AI brings politics closer to voters. Voters will be listened to, more protected and closer to their candidates," Helcer said.

A new robotic system developed by UPM researchers and AURA Innovative Robotics Company can help diagnose neurodegenerative diseases, such as dementia and Parkinson, through the analysis of eye movements. OSCANN Desk is a non-invasive technology developed by researchers from Universidad Politécnica de Madrid (UPM) and the company AURA Innotive Robotics, led by Cecilia García Cena that with a simple and fast test can provide data about brain functioning through the measurement of eye movements. This new system is in the phase of clinical trial authorized by the Spanish Agency of Medicines and Medical Devices in six Spanish hospitals and, thanks to techniques of imaging processing and machine learning, its results will allow doctors to early diagnose neurodegenerative diseases and carry out customized treatments. The diagnosis process of a neurodegenerative disease takes time since symptoms are complex to assess in the early stages of the disease. Besides, there are symptoms that are common to other neurodegenerative diseases such as tremors.

Dementia is a growing condition, overtaking heart disease, lung cancer and stroke as the leading cause of death in the UK, and Alzheimer's Research UK suggests there are over 209,000 new cases of dementia every year in the UK – roughly equivalent to a new case every three minutes. As life expectancy increases, so the likelihood of people developing dementia has risen – making it a pressing public health concern. Early detection can be difficult, too, as Dr Carol Routledge, director of research at Alzheimer's Research UK, explains. Clinical tools are currently not sensitive enough to diagnose early, and blood tests are "not sufficiently specific" to diagnose the diseases causing dementia. She also points out that PET (positron emission tomography) imaging – a type of brain scanning often used to diagnose dementia – can be prohibitively expensive, making it almost impossible to use as widely as needed.