The interest in complex deep neural networks for computer vision applications is increasing. This leads to the need for improving the interpretable capabilities of these models. Recent explanation methods present visualizations of the relevance of pixels from input images, thus enabling the direct interpretation of properties of the input that lead to a specific output. These methods produce maps of pixel importance, which are commonly evaluated by visual inspection. This means that the effectiveness of an explanation method is assessed based on human expectation instead of actual feature importance. Thus, in this work we propose an objective measure to evaluate the reliability of explanations of deep models. Specifically, our approach is based on changes in the network's outcome resulting from the perturbation of input images in an adversarial way. We present a comparison between widely-known explanation methods using our proposed approach. Finally, we also propose a straightforward application of our approach to clean relevance maps, creating more interpretable maps without any loss in essential explanation (as per our proposed measure).

This thesis contributes to the mathematical foundation of domain adaptation as emerging field in machine learning. In contrast to classical statistical learning, the framework of domain adaptation takes into account deviations between probability distributions in the training and application setting. Domain adaptation applies for a wider range of applications as future samples often follow a distribution that differs from the ones of the training samples. A decisive point is the generality of the assumptions about the similarity of the distributions. Therefore, in this thesis we study domain adaptation problems under as weak similarity assumptions as can be modelled by finitely many moments.

The global artificial intelligence in supply chain market is expected to grow at a CAGR of 45.3% from 2019 to reach $21.8 billion by 2027; wherein, Asia-Pacific region is expected to register fastest CAGR throughout the forecast period. Artificial intelligence has emerged as the most potent technologies over the past few years, that is transitioning the landscape of almost all industry verticals. Although enterprise applications based on AI and machine learning (ML) are still in the nascent stages of development, they are gradually beginning to drive innovation strategies of the business. In the supply chain and logistics industry, artificial intelligence is gaining rapid traction among industry stakeholders. Players operating in the supply chain and logistics industry are increasingly realizing the potential of AI to solve the complexities of running a global logistics network.

Stochastic block models (SBMs) are often used to find assortative community structures in networks, such that the probability of connections within communities is higher than in between communities. However, classic SBMs are not limited to assortative structures. In this study, we discuss the implications of this model-inherent indifference towards assortativity or disassortativity, and show that this characteristic can lead to undesirable outcomes for networks which are presupposedy assortative but which contain a reduced amount of information. To circumvent this issue, we introduce a constrained SBM that imposes strong assortativity constraints, along with efficient algorithmic approaches to solve it. These constraints significantly boost community recovery capabilities in regimes that are close to the information-theoretic threshold. They also permit to identify structurally-different communities in networks representing cerebral-cortex activity regions.

In this paper, we deal with the problem of automatic tag recommendation for painting artworks. Diachronic descriptions containing deviations on the vocabulary used to describe each painting usually occur when the work is done by many experts over time. The objective of this work is to provide a framework that produces a more accurate and homogeneous set of tags for each painting in a large collection. To validate our method we build a model based on a weakly-supervised neural network for over $5{,}300$ paintings with hand-labeled descriptions made by experts for the paintings of the Brazilian painter Candido Portinari. This work takes place with the Portinari Project which started in 1979 intending to recover and catalog the paintings of the Brazilian painter. The Portinari paintings at that time were in private collections and museums spread around the world and thus inaccessible to the public. The descriptions of each painting were made by a large number of collaborators over 40 years as the paintings were recovered and these diachronic descriptions caused deviations on the vocabulary used to describe each painting. Our proposed framework consists of (i) a neural network that receives as input the image of each painting and uses frequent itemsets as possible tags, and (ii) a clustering step in which we group related tags based on the output of the pre-trained classifiers.

Causal inference methods are widely applied in the fields of medicine, policy, and economics. Central to these applications is the estimation of treatment effects to make decisions. Current methods make binary yes-or-no decisions based on the treatment effect of a single outcome dimension. These methods are unable to capture continuous space treatment policies with a measure of intensity. They also lack the capacity to consider the complexity of treatment such as matching candidate treatments with the subject. We propose to formulate the effectiveness of treatment as a parametrizable model, expanding to a multitude of treatment intensities and complexities through the continuous policy treatment function, and the likelihood of matching. Our proposal to decompose treatment effect functions into effectiveness factors presents a framework to model a rich space of actions using causal inference. We utilize deep learning to optimize the desired holistic metric space instead of predicting single-dimensional treatment counterfactual. This approach employs a population-wide effectiveness measure and significantly improves the overall effectiveness of the model. The performance of our algorithms is. demonstrated with experiments. When using generic continuous space treatments and matching architecture, we observe a 41% improvement upon prior art with cost-effectiveness and 68% improvement upon a similar method in the average treatment effect. The algorithms capture subtle variations in treatment space, structures the efficient optimizations techniques, and opens up the arena for many applications.

Transformer-based models have pushed the state of the art in many natural language processing tasks. However, one of their main limitations is the quadratic computational and memory cost of the standard attention mechanism. In this paper, we present a new family of Transformer models, which we call the Extended Transformer Construction (ETC), that allows for significant increases in input sequence length by introducing a new global-local attention mechanism between a global memory and the standard input tokens. We also show that combining global-local attention with relative position encodings allows ETC to handle structured data with ease. Empirical results on the Natural Questions data set show the promise of the approach.

Recent surveys, studies, forecasts and other quantitative assessments of AI highlight the role AI plays in fighting the Coronavirus, the business impact of AI, and what the American public feels about it. UC San Diego Health developed and applied an artificial intelligence algorithm to more than 2,000 lung X-ray images, helping radiologists more quickly identify signs of early pneumonia in Covid-19 patients [Becker's Hospital Review] Mayo Clinic teamed up with the state's health department to create an artificial intelligence-powered tool that can identify zones of greater Covid-19 transmission in southern Minnesota [Becker's Hospital Review] The FluSense model, developed by researchers at University of Massachusetts Amherst, was tested in campus clinic waiting rooms. The AI platform was able to analyze coughing sounds and crowd size collected by the handheld device in real-time, then use that data to accurately predict daily illness rates in each clinic [Becker's Hospital Review] The Rambam Hospital in Haifa, Israel, has begun a clinical trial of Cordio Medical's app-based AI system that analyzes speech to diagnose and remotely monitor Covid-19 patients [VentureBeat] Kentucky-based Baptist Health is using an AI platform from remote-patient-monitoring startup Current Health Ltd. to track about 20 Covid-19 patients [WSJ] AI startup SparkBeyond will assist Argentina in looking at how the country can allow citizens to return to work and minimize economic impact. The platform will use data from the Argentinian ministry of health, which aggregates travel, demographic and employment data for each citizen, then integrates hundreds of external data sources to create a wider picture of the situation. It is an area where any country, even countries as big as China and the United States, will find it challenging to achieve the necessary scale of data--from tens to hundreds of millions of humans--to train machine-learning applications that generate robust insights into health and disease.

Customers of any type of business expect help instantly and access to their services in a growing number of ways. Banks are turning to chatbots to help deal with massive volumes of customer interactions. Conversational banking frees up agents for more complex issues, while the move to app-based and web banking sees customers more used to dealing with digital interfaces, of which chatbots and AI virtual assistants are just the latest step. Established banks and their challenger rivals are all keen to develop a conversational banking strategy. Those that have been experimenting for some years find themselves with key advantages over banks stepping fresh into the conversational customer service arena.

While general purpose chatbots haven't shaped user interfaces as radically as early advocates like Facebook may have hoped, when used in a more targeted capacity, they have shown promise in building closer relationships between consumers and brands and making critical enterprise workflows more streamlined. India's Yellow Messenger operates a conversational AI platform used by companies including Accenture, Flipkart and Grab to communicate with employees and customers. The startup is announcing new funding as they officially launch their chatbot platform stateside. The Bengaluru-based company tells TechCrunch they've closed a $20 million Series B led by Lightspeed Venture Partners. The startup previously raised funding from Lightspeed India Partners, which led the startup's Series A last year.