In this thesis, we explore and apply methods inspired by the free energy principle to two important areas in machine learning and neuroscience. The free energy principle is a general mathematical theory of the necessary information-theoretic behaviours of systems which maintain a separation from their environment. A core postulate of the theory is that complex systems can be seen as performing variational Bayesian inference and minimizing an information-theoretic quantity called the variational free energy. The free energy principle originated in, and has been extremely influential in theoretical neuroscience, having spawned a number of neurophysiologically realistic process theories, and maintaining close links with Bayesian Brain viewpoints. The thesis is split into three main parts where we apply methods and insights from the free energy principle to understand questions first in perception, then action, and finally learning. Specifically, in the first section, we focus on the theory of predictive coding, a neurobiologically plausible process theory derived from the free energy principle under certain assumptions, which argues that the primary function of the brain is to minimize prediction errors. We focus on scaling up predictive coding architectures and simulate large-scale predictive coding networks for perception on machine learning benchmarks; we investigate predictive coding's relationship to other classical filtering algorithms, and we demonstrate that many biologically implausible aspects of current models of predictive coding can be relaxed without unduly harming the performance of predictive coding models which allows for a potentially more literal translation of predictive coding theory into cortical microcircuits. In the second part of the thesis, we focus on the application of methods deriving from the free energy principle to action. We study the extension of methods of'active inference', a neurobiologically grounded account of action through variational message passing, to utilize deep artificial neural networks, allowing these methods to'scale up' to be competitive with state of the art deep reinforcement learning methods.

Graph Networks (GNs) enable the fusion of prior knowledge and relational reasoning with flexible function approximations. In this work, a general GN-based model is proposed which takes full advantage of the relational modeling capabilities of GNs and extends these to probabilistic modeling with Variational Bayes (VB). To that end, we combine complementary pre-existing approaches on VB for graph data and propose an approach that relies on graph-structured latent and conditioning variables. It is demonstrated that Neural Processes can also be viewed through the lens of the proposed model. We show applications on the problem of structured probability density modeling for simulated and real wind farm monitoring data, as well as on the meta-learning of simulated Gaussian Process data. We release the source code, along with the simulated datasets.

Researchers at the Technion - Israel Institute of Technology have developed a new way of diagnosing tuberculosis cases, according to a statement.The novel method can diagnose the disease by means of a sticker patch that catches compounds released by the skin, using artificial intelligence to analyze them - resulting in a quick, non-invasive diagnosis.Their findings were published in the medical journal Advanced Science.The WHO's annual TB report found that tuberculosis killed some 1.4 million people in 2019, not much less than the 1.5 million deaths it caused in 2018. The report warned that many countries are not on track to meet targets for successfully diagnosing and treating cases to stop the disease's spread amid the coronavirus pandemic.Before the COVID-19 pandemic, the WHO's report said, many countries had been making steady progress against TB, with a 9% reduction in incidence seen between 2015 and 2019 and a 14% drop in deaths during the same period."Early What makes matters worse is that currently existing diagnosis methods are slow, and at times too expensive or complex for resource-limited settings," Technion explained. "For example, a sputum smear ($2.60 to $10.50 per examination) is too expensive in a location where people live on $1/day, while a mycobacterial culture test takes 4–8 weeks and at least three visits by the patient to finalize the diagnosis and begin treatment." "none";}The device, termed an A-patch, is already in its clinical trial period and is a sought-after diagnostic tool.

The increasing population of elderly people is associated with the need to meet their increasing requirements and to provide solutions that can improve their quality of life in a smart home. In addition to fear and anxiety towards interfacing with systems; cognitive disabilities, weakened memory, disorganized behavior and even physical limitations are some of the problems that elderly people tend to face with increasing age. The essence of providing technology-based solutions to address these needs of elderly people and to create smart and assisted living spaces for the elderly; lies in developing systems that can adapt by addressing their diversity and can augment their performances in the context of their day to day goals. Therefore, this work proposes a framework for development of a Personalized Intelligent Assistant to help elderly people perform Activities of Daily Living (ADLs) in a smart and connected Internet of Things (IoT) based environment. This Personalized Intelligent Assistant can analyze different tasks performed by the user and recommend activities by considering their daily routine, current affective state and the underlining user experience. To uphold the efficacy of this proposed framework, it has been tested on a couple of datasets for modelling an average user and a specific user respectively. The results presented show that the model achieves a performance accuracy of 73.12% when modelling a specific user, which is considerably higher than its performance while modelling an average user, this upholds the relevance for development and implementation of this proposed framework.

Cluster randomized trials (CRTs) randomly assign an intervention to groups of individuals (e.g., clinics or communities), and measure outcomes on individuals in those groups. While offering many advantages, this experimental design introduces challenges that are only partially addressed by existing analytic approaches. First, outcomes are often missing for some individuals within clusters. Failing to appropriately adjust for differential outcome measurement can result in biased estimates and inference. Second, CRTs often randomize limited numbers of clusters, resulting in chance imbalances on baseline outcome predictors between arms. Failing to adaptively adjust for these imbalances and other predictive covariates can result in efficiency losses. To address these methodological gaps, we propose and evaluate a novel two-stage targeted minimum loss-based estimator (TMLE) to adjust for baseline covariates in a manner that optimizes precision, after controlling for baseline and post-baseline causes of missing outcomes. Finite sample simulations illustrate that our approach can nearly eliminate bias due to differential outcome measurement, while other common CRT estimators yield misleading results and inferences. Application to real data from the SEARCH community randomized trial demonstrates the gains in efficiency afforded through adaptive adjustment for cluster-level covariates, after controlling for missingness on individual-level outcomes.

Current model-based reinforcement learning methods struggle when operating from complex visual scenes due to their inability to prioritize task-relevant features. To mitigate this problem, we propose learning Task Informed Abstractions (TIA) that explicitly separates reward-correlated visual features from distractors. For learning TIA, we introduce the formalism of Task Informed MDP (TiMDP) that is realized by training two models that learn visual features via cooperative reconstruction, but one model is adversarially dissociated from the reward signal. Empirical evaluation shows that TIA leads to significant performance gains over state-of-the-art methods on many visual control tasks where natural and unconstrained visual distractions pose a formidable challenge.

Conversational machine reading (CMR) tools allow users to give a description of their scenario and pose a question to them [1] [2]. The CMR tool then processes the rule text in relation to the user scenario and question and either picks an appropriate answer from the set of possible answers A {Yes, No, Irrelevant} or chooses to seek futher clarification before giving an answer from the set A [3]. A number of systems [2] [3] [4] [5] have been developed with a goal to improve the precision of the answers given to the user. However, all the existing tools apply supervised learning technique which require manually labeled dataset. For every new rule text, the supervised techniques will require that a labeled dataset be created. The task of manually labeling dataset is tedious and error prone [6].

ZzappMalaria, a Jerusalem-based startup whose mobile app aims to help identify potential sources of malaria, has won a first prize of $3 million in the IBM Watson AI XPRIZE competition. The firm was also selected as the "Most Inspiring Team" in the People's Choice Award. The IBM Watson AI XPRIZE Challenge was launched in 2016 to promote the use of AI to solve the world's most pressing problems. Aifred Health, a Montreal-based digital health company focused on providing support for clinical decisions for mental health, won second place, getting a $1 million prize. Marinus Analytics, a Pittsburg, US-based firm that uses AI to quickly turn big data into actionable intelligence, helps fight human trafficking by saving hours and sometimes days of investigative time to find traffickers and recover victims.

U.S. fighter jets dropped bombs on Iranian-backed militias in Iraq and Syria. The strike was in response to Iranian-backed militias firing armed drones against U.S. troops in Iraq, which was a response to a U.S. attack in February, which was a response to a militia attack days earlier. A Pentagon spokesman justified the most recent U.S. airstrikes as "necessary, appropriate, and deliberate action designed to limit the risk of escalation--but also to send a clear and unambiguous deterrent message." This may be true, but similar statements have followed similar strikes for years, even decades; yet counter-attacks nonetheless follow (the "deterrent message" doesn't get through), and so it's possible that we are heightening the "risk of escalation," not limiting it. President Joe Biden finds himself in a jam.

Suzuki Motor Corp. warned that the supply of semiconductors and batteries may be tight into the foreseeable future as automakers shift toward electric vehicles, and is planning to increase its stockpiles as a result. "We will be using many batteries and various chips, more rapidly," Osamu Honda, Suzuki's representative director, said at a shareholder meeting on Friday. "We expect chips, batteries and others to always be in tight supply in that case." A global shortage of chips sparked by a range of factors including the coronavirus pandemic, a factory fire in Japan and frigid weather in the U.S. has limited production of everything from cars to game consoles and sent governments scrambling to bolster domestic supply. Japan, which heavily relies on imports, is now eyeing investing trillions of yen to revive its semiconductor manufacturing industry.