Researchers at the MIT Media Lab questioned people around the world about how driverless cars should be programmed to react if their brakes fail. There were stark cultural differences, according to the website Quartz's summary of the research. When forced to choose who survives an accident, people in car-happy Western countries were lukewarm about keeping pedestrians alive at the expense of passengers; they also strongly preferred -- as did people in African countries such as Kenya and South Africa -- to save children over old people. People in Eastern countries, such as China, strongly preferred to save elders over the young and pedestrians over car passengers.

The digital telecommunications receiver is an important context for inference methodology, the key objective being to minimize the expected loss function in recovering the transmitted information. For that criterion, the optimal decision is the Bayesian minimum-risk estimator. However, the computational load of the Bayesian estimator is often prohibitive and, hence, efficient computational schemes are required. The design of novel schemes, striking new balances between accuracy and computational load, is the primary concern of this thesis. Two popular techniques, one exact and one approximate, will be studied. The exact scheme is a recursive one, namely the generalized distributive law (GDL), whose purpose is to distribute all operators across the conditionally independent (CI) factors of the joint model, so as to reduce the total number of operators required. In a novel theorem derived in this thesis, GDL, if applicable, will be shown to guarantee such a reduction in all cases. An associated lemma also quantifies this reduction. For practical use, two novel algorithms, namely the no-longer-needed (NLN) algorithm and the generalized form of the Markovian Forward-Backward (FB) algorithm, recursively factorizes and computes the CI factors of an arbitrary model, respectively. The approximate scheme is an iterative one, namely the Variational Bayes (VB) approximation, whose purpose is to find the independent (i.e. zero-order Markov) model closest to the true joint model in the minimum Kullback-Leibler divergence (KLD) sense. Despite being computationally efficient, this naive mean field approximation confers only modest performance for highly correlated models. A novel approximation, namely Transformed Variational Bayes (TVB), will be designed in the thesis in order to relax the zero-order constraint in the VB approximation, further reducing the KLD of the optimal approximation.

Conventional research attributes the improvements of generalization ability of deep neural networks either to powerful optimizers or the new network design. Different from them, in this paper, we aim to link the generalization ability of a deep network to optimizing a new objective function. To this end, we propose a \textit{nonlinear collaborative scheme} for deep network training, with the key technique as combining different loss functions in a nonlinear manner. We find that after adaptively tuning the weights of different loss functions, the proposed objective function can efficiently guide the optimization process. What is more, we demonstrate that, from the mathematical perspective, the nonlinear collaborative scheme can lead to (i) smaller KL divergence with respect to optimal solutions; (ii) data-driven stochastic gradient descent; (iii) tighter PAC-Bayes bound. We also prove that its advantage can be strengthened by nonlinearity increasing. To some extent, we bridge the gap between learning (i.e., minimizing the new objective function) and generalization (i.e., minimizing a PAC-Bayes bound) in the new scheme. We also interpret our findings through the experiments on Residual Networks and DenseNet, showing that our new scheme performs superior to single-loss and multi-loss schemes no matter with randomization or not.

Adversarial examples can be defined as inputs to a model which induce a mistake - where the model output is different than that of an oracle, perhaps in surprising or malicious ways. Original models of adversarial attacks are primarily studied in the context of classification and computer vision tasks. While several attacks have been proposed in natural language processing (NLP) settings, they often vary in defining the parameters of an attack and what a successful attack would look like. The goal of this work is to propose a unifying model of adversarial examples suitable for NLP tasks in both generative and classification settings. We define the notion of adversarial gain: based in control theory, it is a measure of the change in the output of a system relative to the perturbation of the input (caused by the so-called adversary) presented to the learner. This definition, as we show, can be used under different feature spaces and distance conditions to determine attack or defense effectiveness across different intuitive manifolds. This notion of adversarial gain not only provides a useful way for evaluating adversaries and defenses, but can act as a building block for future work in robustness under adversaries due to its rooted nature in stability and manifold theory.

Explaining the unreasonable effectiveness of deep learning has eluded researchers around the globe. Various authors have described multiple metrics to evaluate the capacity of deep architectures. In this paper, we allude to the radius margin bounds described for a support vector machine (SVM) with hinge loss, apply the same to the deep feed-forward architectures and derive the Vapnik-Chervonenkis (VC) bounds which are different from the earlier bounds proposed in terms of number of weights of the network. In doing so, we also relate the effectiveness of techniques like Dropout and Dropconnect in bringing down the capacity of the network. Finally, we describe the effect of maximizing the input as well as the output margin to achieve an input noise-robust deep architecture.

China now has seminars to teach other countries how to censor free speech as its'techno-dystopia' spreads, a worrying report has found. Governments worldwide are stepping up use of online tools to suppress dissent and tighten their grip on power, a human rights watchdog study found. Chinese officials have held sessions on controlling information with 36 of the 65 countries assessed, and provided telecom and surveillance equipment to a number of foreign governments, researchers said. India led the world in the number of internet shutdowns, with over 100 reported incidents in 2018 so far, claiming that the moves were needed to halt the flow of disinformation and incitement to violence. Many governments, including Saudi Arabia, are employing'troll armies' to manipulate social media and in many cases drown out the voices of dissidents.

WhatsApp is working on a new feature to stop nosy friends scrolling through your messages, according to a new report. The firm will do this by adding Touch ID and Face ID authentication to the messaging app. The Touch ID is set to be available on iPhones using iOS 8 or more while the Face ID will be available for iPhone X, iPhone Xr, iPhone Xs and Xs Max. WhatsApp is working on a new feature to stop nosy friends scrolling through your messages. According to wabetainfo the new feature, which is still under development, will be under Privacy Settings, under'Touch ID'.

PayPal co-founder and prominent Donald Trump donor Peter Thiel has addressed reports that he uses radical life extension therapies that involve blood transfusions using youthful donors. Speaking at the New York Times Dealbook conference Mr Thiel said: "I want to publicly tell you that I'm not a vampire. On the record, I am not a vampire." When questioned further on the matter, he denied that he has ever injected himself with a "young person's blood," a practice that some believe can have a regenerative effect on older people. The I.F.O. is fuelled by eight electric engines, which is able to push the flying object to an estimated top speed of about 120mph.

The new products just revealed by Apple at a special event in Brooklyn this week have several things in common. But most notable is the input of Apple's Chief of Design, Sir Jonathan Ive, universally referred to as Jony. He is involved in new products across Apple, including radical upgrades of favourites such as the MacBook Air and the iPad Pro. Ive has been at Apple since 1992 and his keen eye has been part of the iMac, the iPhone, iPad and Apple Watch. He has just been awarded the 2018 Professor Hawking Fellowship because of what the committee felt included a "remarkable role in championing elegant and innovative design".

The development of so-called neurostimulators may lead to dystopian scenarios whereby hackers create false memories and implant them in people's brains, researchers have warned. The human brain is vulnerable to manipulation through implantable medical devices used to treat things like Parkinson's, according to a practical and theoretical review of this and other scenarios undertaken by the University of Oxford Functional Neurosurgery Group and Russian cyber security firm Kaspersky. Within a decade, technology will also have progressed to the point that commercial memory boosting implants will be available to buy, according to the researchers, while 20 years from now could see a time when it will the technology will be advanced enough to allow for "extensive control over memories." The development of these technologies will have a number of healthcare benefits and will open up the possibility of new bio-connected technologies like increased brain capacity, however it also holds the potential for exploitation and abuse. "New threats resulting from this could include the mass manipulation of groups through implanted or erased memories of political events or conflicts; while'repurposed' cyberthreats could target new opportunities for cyber-espionage or the theft, deletion of, or'locking' of memories (for example, in return for a ransom)," the researchers wrote in their report'The Memory Market: Preparing for a future where cyberthreats target your past'.