Quantum mechanics is a basis for understanding physical phenomena on an atomic scale. An electron point particle of rest mass $m_0$, charge magnitude $e$, and quantized spin magnitude $\hslash /2$, can behave as a wave.

In previous chapters, we encountered the fundamental principles of quantum theory and we saw how the representation of physical magnitudes by Hilbert space operators allows us to construct probabilities for the outcome of any experiment. However, these principles do not tell us what the fundamental physical systems are, how to construct their associated Hilbert spaces, and which operators correspond to physical magnitudes.

In this section, we will analyze transitions between energy eigenstates caused by a transient external force, for example, an EM pulse. We will assume that the external force is weak so that we can describe these decays by a version of perturbation theory. This results into very simple expressions for the transition probabilities and rates with universal validity, that is, they apply to all kinds of phenomena from atomic to nuclear and high-energy physics.

In this chapter we focus on post-training detection of backdoor attacks which replace a patch of pixels by a common backdoor pattern. We focus on scene-plausible perceptible backdoor patterns. Scene-plausibility is important for a perceptible attack to be evasive to human and machine-based detection, whether the attack is physical or digital. Though the focus is on image classification, the methodology could be applied to audio, where for “scene-plausibility” the backdoor pattern does not sound artificial or incongruous, amongst other sounds in the audio clip. For the Neural Cleanse method, the common backdoor pattern may be scene plausible or incongruous. In the latter case, backdoor trigger images (at test time) might be noticed by a human, thus thwarting the attack. The focus here is on defending against patch attacks that are scene plausible, meaning that the backdoor pattern cannot in general be embedded into the same location in every (poisoned) image. For example, a rainbow (one of the attack patterns) must be embedded in the sky (and this location may vary). The main method described builds on RED. It exploits the need for scene-plausibility, and attack “durability,” that the backdoor trigger will be effective in the presence of noise and occlusion.

It is possible to engineer properties of materials, devices, and systems by changing experimentally available control parameters to optimally approach a specific objective. The following sections demonstrate some potential applications of quantum engineering and show how this may be achieved by the development of efficient physical models combined with optimization algorithms.

In this chapter we provide an introduction to deep learning. This includes introducing pattern recognition concepts, neural network architectures, basic optimization techniques (as used by gradient-based deep learning algorithms), and various deep learning paradigms, for example for coping with limited available labeled training data and for improving embedded deep feature representation power. Some of the former include semi-supervised learning, transfer learning, and contrastive learning. Some of the latter include mainstays of deep learning such as convolutional layers, pooling layers, ReLU activations, dropout layers, attention mechanisms, and transformers. Gated recurrent neural networks (such as LSTMs) are not discussed in depth because they are not used in subsequent chapters. Some topics introduced in this chapter, such as neural network inversion and robust classifier training strategies, will be revisited frequently in subsequent chapters, as they form the basis both for attacks against deep learning and for defenses against such attacks.

This chapter argues that subjects originate internally within VP, and from there raise to spec-TP via an operation known as A Movement. The chapter begins (Module 2.1) by outlining the VP Internal Subject Hypothesis, and discussing evidence for it. Module 2.2 considers predicates, arguments, theta roles and theta marking. Module 2.3 turns to explore the syntax of unaccusative subjects, and compares unaccusatives with other type of predicate. Module 2.4 goes on to look at Passivisation, contrasting short/clause-internal and long/cross-clausal passives, and discussing constraints on Passivisation. Module 2.5 looks at the syntax of Raising structures, and compares them with Control structures, establishing criteria for determining whether a given item is a Raising or Control predicate. Module 2.6 notes that many sentences are mixed structures, containing more than one type of predicate (e.g. a passive and an unaccusative predicate). The chapter concludes with a Summary (Module 2.7), Bibliography (Module 2.8), and Workbook (Module 2.9), with some Workbook exercise examples designed for self-study, and others for assignments/seminar discussion.

This chapter discusses how constituents come to be dropped in abbreviated registers of English. It begins (Module 7.1) by examining Subject Drop in colloquial English sentences like ‘<I> can’t find it’, discussing whether they involve Truncation of the periphery above SUBJP. Module 7.2 then goes on to look at Auxiliary+Subject Drop in sentences like ‘<Are you> doing anything tonight?’ and considers whether this results from Weak Syllable Drop in the phonology. Next Module 7.3 considers Article Drop in newspaper headlines, and whether this results from syntactic Truncation, or Article Drop. Module 7.4 goes on to look at omission of Be in newspaper headlines (and its correlation with Article Drop), asking if this involves Truncation, or a tense/agreement deficit. Subsequently Module 7.5 examines Object Drop in product labels (like ‘Don’t stir <it>’), arguing against Topic Drop and in favour of pro-drop. The chapter concludes with a summary (Module 7.6), Bibliography (Module 7.7), and Workbook (Module 7.8), with some Workbook exercise examples designed for self-study, and others for assignments/seminar discussion.

The simplest quantum systems correspond to the smallest nontrivial Hilbert space ${ℂ}^2$. They are called two-level systems. Some physical magnitudes, such as photon polarization or electron spin, are naturally described by the Hilbert space ${ℂ}^2$. However, usually, two-level systems are approximations to more complex systems. Consider, for example, an atom with a lowest-energy state $|0〉$.

Many experiments require the execution of several measurements in a microscopic system. For example, consider a particle A decaying into a pair of particles B and C that move in different directions. Each product particle is detected by a different apparatus at different moments of time. We need a rule that tells us how to incorporate information obtained from the first measurement in order to predict the outcome of the second.

One of the most important motives for the development of quantum theory was the need to understand the structure of atoms and molecules, and to account for their emission spectra. Later on, analogous issues were raised for other composite systems, such as nuclei and hadrons. In all cases, the answer requires finding the discrete spectrum of the Hamiltonian $\widehat{H}$ that describes the composite system, that is, solving the eigenvalue equation $\widehat{H}|n〉=E_n|n〉$ for the Hamiltonian.

This chapter provides an overview of sexuality and its role in romantic relationships. It starts with a discussion of the commonalities and diversity in people’s sexual desires, sexual attitudes, motives for sex, and sexual behaviors. It then covers how often people have sex and the relational implications of more or less frequent sex. Next, this chapter reviews predictors of sexual satisfaction and the relation between sexual satisfaction and relational satisfaction, more broadly. Finally, the chapter ends with a review of how people initiate sex in both casual contexts and in ongoing romantic relationships.