搜索结果：exempt

Surprisal Theory is often characterized as a computational-level explanation per (Marr, 1982). We argue in this work that, even though a computational level narrative has been used to support "representation-agnostic research" within computational psycholinguistics, the movement toward black box systems embodied by large language models (LLMs) does not exempt modelers using the surprisal metric from the representational decisions required by computational-level characterizations. In fact, we argue that the uncritical use of LLM-surprisal obfuscates the representational and algorithmic-level commitments of different models. In three analyses, we show that the choice of algorithm and model architecture play significant roles in the computation of language model probabilities. We advise that researchers who wish to test Surprisal Theory re-evaluate the practice of treating large language model probabilities as interchangeable

Over the past few years an increasing number of states in the US have adopted new privacy laws. The majority of these laws require compliance with universal opt-out mechanisms (UOOMs), which allow consumers to send legally binding opt-out signals. However, a number of laws generally do not allow UOOMs to be enabled by default. While some laws exempt privacy-protective software from this prohibition, the exemption does not apply to pre-installed software, e.g., a privacy-protective web browser bundled with an operating system. The reason for not allowing default opt-out settings for pre-installed software is to ensure that settings reflect consumers' "affirmative, freely given, and unambiguous choice," as, for example, the Colorado Privacy Act (CPA) is putting it. However, prohibiting vendors of privacy-protective software from turning on UOOMs by default can force them into committing unfair or deceptive acts or practices under the FTC Act and equivalent state laws. Thus, whether UOOMs can be turned on by default on pre-installed software should depend on consumers' privacy expectations. For pre-installed software that is creating a reasonable expectation for consumers that their p

The importance of simulating pinning arrays in superconducting samples for the increase of critical currents has been increasing in the last few years. Since the Time Dependent Ginzburg Landau (TDGL) can be more accurate than alternative methodologies, the simulation procedures involving it are critical to design devices that can sustain higher critical currents and, therefore, to the field of applied superconductivity. In this article, a simple novel algorithm is presented for the reduction of bias and optimisation of execution time in iterative time dependent simulations, applied to TDGL solutions of superconducting samples. Taking a time series approach to the magnetic response of the sample, stationary solutions are found for each step in the evolution of the applied external field, leading to bias reduction and minimisation of iterations needed to be spent at each step in the applied field. The results are presented for a pure superconductor, in a framework of simulations via link variable technique, with simple Euler algorithm for the solution in time, but the implementation can be adapted easily to deal with adaptive step size solutions or semi-implicit methods, which are no

The intersection of artificial intelligence (AI) and digital forensics (DF) is becoming increasingly complex, ubiquitous, and pervasive, with overlapping techniques and technologies being adopted in all types of scientific and technical inquiry. Despite incredible advances, forensic sciences are not exempt from errors and remain vulnerable to fallibility. To mitigate the limitations of errors in DF, the systemic complexity is identified and addressed with the adoption of human-readable artifacts and open standards. A DF AI model schema based on the state of the art is outlined.

Physics is a model of nature able to both describe and predict the results of measurements made with respect to reference systems. These reference systems, in turn, are themselves physical and thus subject to the laws of physics. The situation is no different when the model in use is quantum mechanics: states and observables are relative entities, and reference frames are not exempt from exhibiting quantum behavior. In recent years, the scientific community has shown renewed interest in quantum reference frames, particularly in connection with the covariance of physical laws and quantum resources. However, current approaches fall short of providing a complete prescription for predicting observables associated solely with degrees of freedom accessible from the quantum reference frame. In pursuit of such a description, we show that while this is fully feasible for two-particle systems, there are irreducible difficulties that arise in many-body systems. In particular, within the framework of Galilean relativity, with absolute time, we demonstrate that a canonical and relational description with respect to a particle in the system cannot be achieved through any unitary transformation.

We model the impact of local vaccine mandates on the spread of vaccine-preventable infectious diseases, which in the absence of vaccines will mainly affect children. Examples of such diseases are measles, rubella, mumps and pertussis. To model the spread of the pathogen, we use a stochastic SIR (Susceptible, Infectious, Recovered) model with two levels of mixing in a closed population, often referred to as the household model. In this model individuals make local contacts within a specific small subgroup of the population (e.g.\ within a household or a school class), while they also make global contacts with random people in the population at a much lower rate than the rate of local contacts. We consider what happens if schools are given freedom to impose vaccine mandates on all of their pupils, except for the pupils that are exempt from vaccination because of medical reasons. We investigate how such a mandate affects the probability of an outbreak of a disease and the probability that a pupil that is medically exempt from vaccination, gets infected during an outbreak. We show that if the population vaccine coverage is close to the herd-immunity level then both probabilities may in

Diffraction causes waves to spread out as they propagate freely. The tighter the lateral confinement, the faster the spreading. Past research on how to suppress diffraction has been based on wave engineering and has led so far to idealized waves that, in real settings, eventually diffract. Here, we find a propagating light wave structure naturally present in optical vortices, a natural skyrmion, that is exempt from diffraction. Moreover, diffraction-free propagation occurs with lateral confinement at any scale below the wavelength of light. In our experiments, we observe non-diffraction over a propagation distance above three orders of magnitude greater than expected from the skyrmion subwavelength size. We thus provide a factual, real-world form of ideal non-diffracting propagation. This form substantially differs from previous forms of light propagation, including propagating optical skyrmions known to date, and could open up new perspectives in its various applications.

Recent antitrust regulations in several countries have granted exemptions for collusion aimed at achieving environmental goals. Firms can apply for exemptions if collusion helps to develop or to implement costly clean technology, particularly in sectors like renewable energy, where capital costs are high and economies of scale are significant. However, if the cost of the green transition is unknown to the competition regulator, firms might exploit the exemption by fixing prices higher than necessary. The regulator faces the decision of whether to permit collusion and whether to commission an investigation of potential price fixing, which incurs costs. We fully characterise the equilibria in this scenario that depend on the regulator's belief about the high cost of green transition. If the belief is high enough, collusion will be allowed. We also identify conditions under which a regulator's commitment to always investigate price fixing is preferable to making discretionary decisions.

This paper presents an a priori error analysis of the Deep Mixed Residual method (MIM) for solving high-order elliptic equations with non-homogeneous boundary conditions, including Dirichlet, Neumann, and Robin conditions. We examine MIM with two types of loss functions, referred to as first-order and second-order least squares systems. By providing boundedness and coercivity analysis, we leverage Céa's Lemma to decompose the total error into the approximation, generalization, and optimization errors. Utilizing the Barron space theory and Rademacher complexity, an a priori error is derived regarding the training samples and network size that are exempt from the curse of dimensionality. Our results reveal that MIM significantly reduces the regularity requirements for activation functions compared to the deep Ritz method, implying the effectiveness of MIM in solving high-order equations.

While topology can impose obstructions to exponentially localized Wannier functions, certain topological insulators are exempt from such Wannier obstructions. The absence of the Wannier obstructions can further accompany topological boundary states that are detachable from the bulk bands. Here, we elucidate a close connection between these detachable topological boundary states and non-Hermitian topology. Identifying topological boundary states as non-Hermitian topology, we demonstrate that intrinsic non-Hermitian topology leads to the inevitable spectral flow. By contrast, we show that extrinsic non-Hermitian topology underlies the detachment of topological boundary states and clarify anti-Hermitian topology of the detached boundary states. Based on this connection and $K$-theory, we complete the tenfold classification of Wannier localizability and detachable topological boundary states.

A few years ago, Kurzweil and Hazak developed the Configurationally Resolved Super-Transition-Arrays (CRSTA) method for the computation of hot-plasma radiative opacity. Their approach, based on a temporal integration, is an important refinement of the standard Super-Transition-Arrays (STA) approach, which enables one to recover the underlying structure of the STAs, made of unresolved transition arrays. The CRSTA formalism relies on the use of complex pseudo partition functions, depending on the considered one-electron jump. In this article, we find that, despite the imaginary part, the doubly-recursive relation which was introduced in the original STA method to avoid problems due to alternating-sign terms in partition functions, is still applicable, robust, efficient, and exempt of numerical instabilities. This was rather unexpected, in particular because of the occurrence of trigonometric functions, or Chebyshev polynomials, which can be either positive or negative. We also show that, in the complex case, the recursion relation can be presented in a form where the vector of real and imaginary parts at a given iteration is therefore obtained by a sum of the rotated previous ones.