Search

Our study addresses the challenges of building datasets to understand the risks associated with organized activities and human trafficking through commercial sex advertisements. These challenges include data scarcity, rapid obsolescence, and privacy concerns. Traditional approaches, which are not automated and are difficult to reproduce, fall short in addressing these issues. We have developed a reproducible and automated methodology to analyze five million advertisements. In the process, we identified further challenges in dataset creation within this sensitive domain. This paper presents a streamlined methodology to assist researchers in constructing effective datasets for combating organized crime, allowing them to focus on advancing detection technologies., Comment: LXAI Workshop at the 2024 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2024)

A Planck scale inflationary era -- in a quantum gravity theory predicting discreteness of quantum geometry at the fundamental scale -- produces the scale invariant spectrum of inhomogeneities with very small tensor-to-scalar ratio of perturbations and a hot big bang leading to a natural dark matter genesis scenario. Here we evoke the possibility that some of the major puzzles in cosmology would have an explanation rooted in quantum gravity.

This project tackles the pressing issue of human trafficking in online C2C marketplaces through advanced Natural Language Processing (NLP) techniques. We introduce a novel methodology for generating pseudo-labeled datasets with minimal supervision, serving as a rich resource for training state-of-the-art NLP models. Focusing on tasks like Human Trafficking Risk Prediction (HTRP) and Organized Activity Detection (OAD), we employ cutting-edge Transformer models for analysis. A key contribution is the implementation of an interpretability framework using Integrated Gradients, providing explainable insights crucial for law enforcement. This work not only fills a critical gap in the literature but also offers a scalable, machine learning-driven approach to combat human exploitation online. It serves as a foundation for future research and practical applications, emphasizing the role of machine learning in addressing complex social issues.

Electron tomography has become a commonly used tool to investigate the three-dimensional (3D) structure of nanomaterials, including colloidal nanoparticle assemblies. However, electron microscopy is typically carried out under high vacuum conditions. Therefore, pre-treatment sample preparation is needed for assemblies obtained by (wet) colloid chemistry methods, including solvent evaporation and deposition on a solid TEM support. As a result of this procedure, changes are consistently imposed on the actual nanoparticle organization. Therefore, we propose herein the application of electron tomography of nanoparticle assemblies while in their original colloidal liquid environment. To address the challenges related to electron tomography in liquid, we devised a method that combines fast data acquisition in a commercial liquid-TEM cell, with a dedicated alignment and reconstruction workflow. We present the application of this method to two different systems, which exemplify the difference between conventional and liquid tomography, depending on the nature of the protecting ligands. 3D reconstructions of assemblies comprising polystyrene-capped Au nanoparticles encapsulated in polymeric shells revealed less compact and more distorted configurations for experiments performed in a liquid medium compared to their dried counterparts. On the other hand, quantitative analysis of the surface-to-surface distance of self-assembled Au nanorods in water agrees with previously reported dimensions of the ligand layers surrounding the nanorods, which are in much closer contact when in similar but dried assemblies. This study, therefore, emphasizes the importance of developing high-resolution characterization tools that preserve the native environment of colloidal nanostructures., Comment: 32 pages, 12 figures, 2 tables, submitted

A brief overview of the discovery that macroscopic black holes are thermodynamical systems is presented. They satisfy the laws of thermodynamics and are associated with a temperature and an entropy equal to one quarter of their horizon area in Planck units. They emit black body radiation and slowly evaporate as a consequence of Heisenberg's uncertainty principle. The problem of understanding the microscopic source of their large entropy, as well as the nature of their final fate after evaporation, are discussed from the perspective of approaches to quantum gravity that predict discreteness at the Planck scale. We review encouraging first steps in computing black hole entropy and briefly discuss their implications for the black hole information puzzle., Comment: Prepared for the Encyclopedia of Mathematical Physics, Elsevier, 2024

In spite or the large astronomical evidence for its existence, the nature of dark matter remains enigmatic. Particles that interact only, or almost only, gravitationally, in particular with masses around the Planck mass -- the fundamental scale in quantum gravity, are intriguing candidates. Here we show that there is a theoretical possibility to directly detect such particles using highly sensitive gravity-mediated quantum phase shifts. In particular, we consider a protocol utilizing Josephson junctions., Comment: A refined protocol using Josephson junctions has been added

The traditional presentation of Unimodular Gravity (UG) consists on indicating that it is an alternative theory of gravity that restricts the generic diffeomorphism invariance of General Relativity. In particular, as often encountered in the literature, unlike General Relativity, Unimodular Gravity is invariant solely under volume-preserving diffeomorphisms. That characterization of UG has led to some confusion and incorrect statements in various treatments on the subject. For instance, sometimes it is claimed (mistakenly) that only spacetime metrics such that $|$det $g_{\mu \nu}| = 1$ can be considered as valid solutions of the theory. Additionally, that same (incorrect) statement is often invoked to argue that some particular gauges (e.g. the Newtonian or synchronous gauge) are not allowed when dealing with cosmological perturbation theory in UG. The present article is devoted to clarify those and other misconceptions regarding the notion of diffeomorphism invariance, in general, and its usage in the context of UG, in particular., Comment: 26 pages, 2 figures, references updated. Version accepted for publication

The modern pervasiveness of large-scale deep neural networks (NNs) is driven by their extraordinary performance on complex problems but is also plagued by their sudden, unexpected, and often catastrophic failures, particularly on challenging scenarios. Existing algorithms that provide risk-awareness to NNs are complex and ad-hoc. Specifically, these methods require significant engineering changes, are often developed only for particular settings, and are not easily composable. Here we present capsa, a framework for extending models with risk-awareness. Capsa provides a methodology for quantifying multiple forms of risk and composing different algorithms together to quantify different risk metrics in parallel. We validate capsa by implementing state-of-the-art uncertainty estimation algorithms within the capsa framework and benchmarking them on complex perception datasets. We demonstrate capsa's ability to easily compose aleatoric uncertainty, epistemic uncertainty, and bias estimation together in a single procedure, and show how this approach provides a comprehensive awareness of NN risk., Comment: Neural Information Processing Systems (NeurIPS) 2022. Workshop on Machine Learning for Autonomous Driving (ML4AD)

We explicitly express the Minkowski vacuum of a massless scalar field in terms of the particle notion associated with suitable spherical conformal killing fields. These fields are orthogonal to the light wavefronts originating from a sphere with a radius of $r_{H}$ in flat spacetime: a bifurcate conformal killing horizon that exhibits semiclassical features similar to those of black hole horizons and Cauchy horizons of spherically symmetric black holes. Our result highlights the quantum aspects of this analogy and extends the well-known decomposition of the Minkowski vacuum in terms of Rindler modes, which are associated with the boost Killing field normal to a pair of null planes in Minkowski spacetime (the basis of the Unruh effect). While some features of our result have been established by Kay and Wald's theorems in the 90s -- on quantum field theory in stationary spacetimes with bifurcate Killing horizons -- the added value we provide here lies in the explicit expression of the vacuum., Comment: Typos corrected

The black hole information puzzle can be resolved if two conditions are met. Firstly, if the information of what falls inside a black hole remains encoded in degrees of freedom that persist after the black hole completely evaporates. These degrees of freedom should be capable of purifying the information. Secondly, if these purifying degrees of freedom do not significantly contribute to the system's energy, as the macroscopic mass of the initial black hole has been radiated away as Hawking radiation to infinity. The presence of microscopic degrees of freedom at the Planck scale provides a natural mechanism for achieving these two conditions without running into the problem of the large pair-creation probabilities of standard remnant scenarios. In the context of Hawking radiation, the first condition implies that correlations between the {\em in} and {\em out} Hawking partner particles need to be transferred to correlations between the {\em microscopic degrees of freedom} and the {\em out} partners in the radiation. This transfer occurs dynamically when the {\em in} partners reach the singularity inside the black hole, entering the UV regime of quantum gravity where the interaction with the microscopic degrees of freedom becomes strong. The second condition suggests that the conventional notion of the vacuum's uniqueness in quantum field theory should fail when considering the full quantum gravity degrees of freedom. In this paper, we demonstrate both key aspects of this mechanism using a solvable toy model of a quantum black hole inspired by loop quantum gravity., Comment: arXiv admin note: text overlap with arXiv:2301.03951, arXiv:1703.09149

In this paper we construct a solvable toy model of the quantum dynamics of the interior of a spherical black hole with falling spherical scalar field excitations. We first argue about how some aspects of the quantum gravity dynamics of realistic black holes emitting Hawking radiation can be modelled using Kantowski-Sachs solutions with a massless scalar field when one focuses on the deep interior region $r\ll M$ (including the singularity). Further, we show that in the $r\ll M$ regime, and in suitable variables, the KS model becomes exactly solvable at both the classical and quantum levels. The quantum dynamics inspired by loop quantum gravity is revisited. We propose a natural polymer-quantization where the area $a$ of the orbits of the rotation group is quantized. The polymer (or loop) dynamics is closely related with the Schroedinger dynamics away from the singularity with a form of continuum limit naturally emerging from the polymer treatment. The Dirac observable associated to the mass is quantized and shown to have an infinite degeneracy associated to the so-called $\epsilon$-sectors. Suitable continuum superpositions of these are well defined distributions in the fundamental Hilbert space and satisfy the continuum Schroedinger dynamics.

Algorithms that involve both forecasting and optimization are at the core of solutions to many difficult real-world problems, such as in supply chains (inventory optimization), traffic, and in the transition towards carbon-free energy generation in battery/load/production scheduling in sustainable energy systems. Typically, in these scenarios we want to solve an optimization problem that depends on unknown future values, which therefore need to be forecast. As both forecasting and optimization are difficult problems in their own right, relatively few research has been done in this area. This paper presents the findings of the ``IEEE-CIS Technical Challenge on Predict+Optimize for Renewable Energy Scheduling," held in 2021. We present a comparison and evaluation of the seven highest-ranked solutions in the competition, to provide researchers with a benchmark problem and to establish the state of the art for this benchmark, with the aim to foster and facilitate research in this area. The competition used data from the Monash Microgrid, as well as weather data and energy market data. It then focused on two main challenges: forecasting renewable energy production and demand, and obtaining an optimal schedule for the activities (lectures) and on-site batteries that lead to the lowest cost of energy. The most accurate forecasts were obtained by gradient-boosted tree and random forest models, and optimization was mostly performed using mixed integer linear and quadratic programming. The winning method predicted different scenarios and optimized over all scenarios jointly using a sample average approximation method.

Non-singular black holes models can be described by modified classical equations motivated by loop quantum gravity. We investigate what happens when the sine function typically used in the modification is replaced by an arbitrary bounded function, a generalization meant to study the effect of ambiguities such as the choice of representation of the holonomy. A number of features can be determined without committing to a specific choice of functions. We find generic singularity resolution. The presence and number of horizons is determined by global features of the function regularizing the angular components of the connection, and the presence and number of bounces by global features of the function regularizing the time component. The trapping or anti-trapping nature of regions inside horizons depends on the relative location with respect to eventual bounces. We use these results to comment on some of the ambiguities of polymer black hole models., Comment: 32 pages, many figures

A discrete space-time structure lying at about the Planck scale may become manifest in the form of very small violations of the conservation of the matter energy-momentum tensor. In order to include such kind of violations, forbidden within the General Relativity framework, the theory of unimodular gravity seems as the simplest option to describe the gravitational interaction. In the cosmological context, a direct consequence of such violation of energy conservation might be heuristically viewed a "diffusion process of matter (both dark and ordinary)" into an effective dark energy term in Einstein's equations, which leads under natural assumptions to an adequate estimate for the value of the cosmological constant. Previous works have also indicated that these kind of models might offer a natural scenario to alleviate the Hubble tension. In this work, we consider a simple model for thecosmological history including a late time occurrence of such energy violation and study the modifications of the predictions for the anisotropy and polarization of the Cosmic Microwave Background (CMB). We compare the model's predictions with recent data from the CMB, Supernovae Type Ia, cosmic chronometers and Baryon Acoustic Oscillations. The results show the potential of this type of model to alleviate the Hubble tension., Comment: 17 pages, 4 figures

We present two alternative perspectives for the resolution of Hawking's information puzzle in black hole evaporation. The two views are deeply contrasting, yet they share several common aspects. One of them is the central role played by the existence of the interior singularity (whose physical relevance is implied by the singularity theorems of Penrose) that we expect to be replaced by a region described by a more fundamental quantum gravity formulation. Both views rely on the notion that the standard effective quantum field theoretic perspective would require some deep modifications. In this respect both of our scenarios are deeply influenced by ideas that Roger Penrose has advocated at various times and thus serves to illustrate the lasting influence that his deep thinking on these and related matters continues to have on the modern thinking about fundamental aspects of both quantum theory and gravitation. Despite that, there is of course no claim that R. Penrose would agree with any of the concrete proposals that will be discussed here., Comment: Prepared for the AVS Quantum Science Special Topic Collection Celebrating Sir Roger Penrose's Nobel Prize

Support vector machines (SVMs) are popular learning algorithms to deal with binary classification problems. They traditionally assume equal misclassification costs for each class; however, real-world problems may have an uneven class distribution. This article introduces EBCS-SVM: evolutionary bilevel cost-sensitive SVMs. EBCS-SVM handles imbalanced classification problems by simultaneously learning the support vectors and optimizing the SVM hyperparameters, which comprise the kernel parameter and misclassification costs. The resulting optimization problem is a bilevel problem, where the lower level determines the support vectors and the upper level the hyperparameters. This optimization problem is solved using an evolutionary algorithm (EA) at the upper level and sequential minimal optimization (SMO) at the lower level. These two methods work in a nested fashion, that is, the optimal support vectors help guide the search of the hyperparameters, and the lower level is initialized based on previous successful solutions. The proposed method is assessed using 70 datasets of imbalanced classification and compared with several state-of-the-art methods. The experimental results, supported by a Bayesian test, provided evidence of the effectiveness of EBCS-SVM when working with highly imbalanced datasets., Comment: Copyright 2022 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works

We show that the quantization ambiguities of loop quantum cosmology, when considered in wider generality, can be used to produce discretionary dynamical behavior. There is an infinite dimensional space of ambiguities which parallels the infinite list of higher curvature corrections in perturbative quantum gravity. There is however an ensemble of qualitative consequences which are generic in the sense that they are independent of these ambiguities. Among these, one has well defined fundamental dynamics across the big bang, and the existence of extra microscopic quantum degrees of freedom that might be relevant in discussions about unitarity in quantum gravity. We show that (in addition to the well known bouncing solutions of the effective equations) there are other generic type of solutions for sufficiently soft initial conditions in the matter sector (tunneling solutions) where the scale factor goes through zero and the spacetime orientation is inverted. We also show that generically, a contracting semiclassical universe branches off at the big bang into a quantum superposition of universes with different quantum numbers. Despite their lack of quantitative predictive power these models offer a fertile playground for the discussion of qualitative and conceptual issues in quantum gravity.

We seek to evaluate the detection performance of a rapid primary screening tool of Covid-19 solely based on the cough sound from 8,380 clinically validated samples with laboratory molecular-test (2,339 Covid-19 positives and 6,041 Covid-19 negatives). Samples were clinically labeled according to the results and severity based on quantitative RT-PCR (qRT-PCR) analysis, cycle threshold, and lymphocytes count from the patients. Our proposed generic method is an algorithm based on Empirical Mode Decomposition (EMD) with subsequent classification based on a tensor of audio features and a deep artificial neural network classifier with convolutional layers called DeepCough'. Two different versions of DeepCough based on the number of tensor dimensions, i.e. DeepCough2D and DeepCough3D, have been investigated. These methods have been deployed in a multi-platform proof-of-concept Web App CoughDetect to administer this test anonymously. Covid-19 recognition results rates achieved a promising AUC (Area Under Curve) of 98.800.83%, sensitivity of 96.431.85%, and specificity of 96.201.74%, and 81.08%5.05% AUC for the recognition of three severity levels. Our proposed web tool and underpinning algorithm for the robust, fast, point-of-need identification of Covid-19 facilitates the rapid detection of the infection. We believe that it has the potential to significantly hamper the Covid-19 pandemic across the world.

The definition of well-behaved coordinate charts for black hole spacetimes can be tricky, as they can lead for example to either unphysical coordinate singularities in the metric (e.g. $r=2M$ in the Schwarzschild black hole) or to an implicit dependence of the chosen coordinate to physical relevant coordinates (e.g. the dependence of the null coordinates in the Kruskal metric). Here we discuss two approaches for coordinate choices in spherical symmetry allowing us to discuss explicitly "solitary" and spherically symmetric black holes from a regular horizon to null infinity. The first approach relies on a construction of a regular null coordinate (where regular is meant as being defined from the horizon to null infinity) given an explicit solution of the Einstein-matter equations. The second approach is based on an affine-null formulation of the Einstein equations and the respective characteristic initial value problem. In particular, we present a derivation of the Reissner-Nordstr\"om black holes expressed in terms of these regular coordinates., Comment: 17 pages

We propose a model of inflation driven by the relaxation of an initially Planckian cosmological constant due to diffusion. The model can generate a (approximately) scale invariant spectrum of (adiabatic) primordial perturbations with the correct amplitudes and red tilt without an inflaton. The inhomogeneities observable in the CMB arise from those associated to the fundamental Planckian granularity that are imprinted into the standard model Higgs scalar fluctuations during the inflationary phase. The process admits a semiclassical interpretation and avoids the trans-Planckian problem of standard inflationary scenarios based on the role of vacuum fluctuations. The deviations from scale invariance observed in the CMB are controlled by the self coupling constant of the Higgs scalar of the standard model of particle physics. The thermal production of primordial black holes can produce the amount of cold dark matter required by observations. For natural initial conditions set at the Planck scale the amplitude and tilt of the power spectrum of perturbations observed at the CMB depend only on known parameters of the standard model such as the self coupling of the Higgs scalar and its mass., Comment: Peer reviewed version

Since the appearance of Einstein's paper {\em"On the Electrodynamics of Moving Bodies"} and the birth of special relativity, it is understood that the theory was basically coded within Maxwell's equations. The celebrated mass-energy equivalence relation, $E=mc^2$, is derived by Einstein using thought experiments involving the kinematics of the emission of light (electromagnetic energy) and the relativity principle. Text book derivations often follow paths similar to Einstein's, or the analysis of the kinematics of particle collisions interpreted from the perspective of different inertial frames. All the same, in such derivations the direct dynamical link with hypothetical fundamental fields describing matter (e.g. Maxwell theory or other) is overshadowed by the use of powerful symmetry arguments, kinematics, and the relativity principle. Here we show that the formula can be derived directly form the dynamical equations of a massless matter model confined in a box (which can be thought of as a toy model of a composite particle). The only assumptions in the derivation are that the field equations hold and the energy-momentum tensor admits a universal interpretation in arbitrary coordinate systems. The mass-energy equivalence relation follows from the inertia or (taking the equivalence principle for granted) weight of confined field radiation. The present derivation offers an interesting pedagogical perspective on the formula providing a simple toy model on the origin of mass and a natural bridge to the foundations of general relativity.

The tension between the value of the Hubble constant $H_0$ determined from local supernovae data and the one inferred from the cosmic microwave background based on the $\Lambda$CDM cosmological model may indicate the need for new physics. Here, we show that this `Hubble tension' can be resolved in models involving an effective energy flux from the matter sector into dark energy resulting naturally from a combination of unimodular gravity and an energy diffusion process. The scheme is one where dark energy has the standard equation of state $w=-1$. This proposal provides an alternative phenomenological paradigm accounting for the observations, while offering a general framework to study diffusion effects coming from novel fundamental physical processes., Comment: 12 pages. v2: Minor changes; v3: minor changes, references added

In approaches to quantum gravity, where smooth spacetime is an emergent approximation of a discrete Planckian fundamental structure, any effective smooth field theoretical description would miss part of the fundamental degrees of freedom and thus break unitarity. This is applicable also to trivial gravitational field (low energy) idealizations realized by the use of the Minkowski background geometry which, as any other spacetime geometry, corresponds, in the fundamental description, to infinitely many different and closely degenerate discrete microstates. The existence of such microstates provides a large reservoir for information to be coded at the end of black hole evaporation and thus opens the way to a natural resolution of the black hole evaporation information puzzle. In this paper we show that these expectations can be made precise in a simple quantum gravity model for cosmology motivated by loop quantum gravity. Concretely, even when the model is fundamentally unitary, when microscopic degrees of freedom irrelevant to low-energy cosmological observers are suitably ignored, pure states in the effective description evolve into mixed states due to decoherence with the Planckian microscopic structure. Moreover, in the relevant physical regime these hidden degrees freedom do not carry any `energy' and thus realize in a fully quantum gravitational context the idea (emphasized before by Unruh and Wald) that decoherence can take place without dissipation, now in a concrete gravitational model strongly motivated by quantum gravity. All this strengthen the perspective of a quite conservative and natural resolution of the black hole evaporation puzzle where information is not destroyed but simply degraded (made unavailable to low energy observers) into correlations with the microscopic structure of the quantum geometry at the Planck scale.

In a recent work we have argued that nosy energy momentum diffusion due to space-time discreteness at the Planck scale (naturally expected to arise from quantum gravity) can be responsible for the generation of a cosmological constant during the electro-weak phase transition era of the cosmic evolution. Simple dimensional analysis and an effectively Brownian description of the propagation of fundamental particles on a granular background yields a cosmological constant of the order of magnitude of the observed value, without fine tuning. While the energy diffusion is negligible for matter in standard astrophysical configurations (from ordinary stars to neutron stars) here we argue that a similar diffusion mechanism could, nonetheless be important for black holes. If such effects are taken into account two observational puzzles might be solved by a single mechanism: the `$H_0$ tension' and the relatively low rotational spin of the black holes detected via gravitational wave astronomy.

In approaches to quantum gravity, where smooth spacetime is an emergent approximation of a discrete Planckian fundamental structure, any standard effective field theoretical description will miss part of the degrees of freedom and thus break unitarity. Here we show that these expectations can be made precise in loop quantum cosmology. Concretely, even when loop quantum cosmology is unitary at the fundamental level, when microscopic degrees of freedom, irrelevant to low-energy cosmological observers, are suitably ignored, pure states in the effective description evolve into mixed states due to decoherence with the Planckian microscopic structure. When extrapolated to black hole formation and evaporation, this concrete example provides a key physical insight for a natural resolution of Hawking's information paradox.

When probed with conformally invariant matter fields, light cones in Minkowski spacetime satisfy thermodynamical relations which are the analog of those satisfied by stationary black holes coupled to standard matter fields. These properties stem from the fact that light cones are conformal Killing horizons stationary with respect to observers following the radial conformal Killing fields in flat spacetime. The four laws of light cone thermodynamics relate notions such as (conformal) temperature, (conformal) surface gravity, (conformal) energy and a conformally invariant notion related to area change. These quantities do not admit a direct physical interpretation in flat spacetime. However, they become the usual thermodynamical quantities when Minkowski is mapped, via a Weyl transformation, to a target spacetime where the conformal Killing field becomes a proper Killing field. In this paper we study the properties of such spacetimes. The simplest realisation turns out to be the Bertotti-Robinson solution, which is known to encode the near horizon geometry of near extremal and extremal charged black holes. The analogy between light cones in flat space and black hole horizons is therefore strengthened. The construction works in arbitrary dimensions; in two dimensions one recovers the Jackiv-Teitelboim black hole of dilaton gravity. Other interesting realisations are also presented., Comment: 23 pages, 7 figures; v2: typos corrected, matches published version

The value of the cosmological constant is explained in terms of a noisy diffusion of energy from the low energy particle physics degrees of freedom to the fundamental Planckian granularity which is expected from general arguments in quantum gravity. The quantitative success of our phenomenological model is encouraging and provides possibly useful insights about physics at the scale of quantum gravity., Comment: Essay written for the Gravity Research Foundation 2018 Awards for Essays on Gravitation. arXiv admin note: text overlap with arXiv:1711.05183

We argue that discreteness at the Planck scale (naturally expected to arise from quantum gravity) might manifest in the form of minute violations of energy-momentum conservation of the matter degrees of freedom when described in terms of (idealized) smooth fields on a smooth spacetime. In the context of applications to cosmology such `energy diffusion' from the low energy matter degrees of freedom to the discrete structures underlying spacetime leads to the emergence of an effective dark energy term in Einstein's equations. We estimate this effect using a (relational) hypothesis about the materialization of discreteness in quantum gravity which is motivated by the strict observational constraints supporting the validity of Lorentz invariance at low energies. The predictions coming from simple dimensional analysis yield a cosmological constant of the order of magnitude of the observed value without fine tuning., Comment: Typos corrected, closer to published version

We show that null surfaces defined by the outgoing and infalling wave fronts emanating from and arriving at a sphere in Minkowski spacetime have thermodynamical properties that are in strict formal correspondence with those of black hole horizons in curved spacetimes. Such null surfaces, made of pieces of light cones, are bifurcate conformal Killing horizons for suitable conformally stationary observers. They can be extremal and non-extremal depending on the radius of the shining sphere. Such conformal Killing horizons have a constant light cone (conformal) temperature, given by the standard expression in terms of the generalisation of surface gravity for conformal Killing horizons. Exchanges of conformally invariant energy across the horizon are described by a first law where entropy changes are given by $1/(4\ell_p^2)$ of the changes of a geometric quantity with the meaning of horizon area in a suitable conformal frame. These conformal horizons satisfy the zeroth to the third laws of thermodynamics in an appropriate way. In the extremal case they become light cones associated with a single event; these have vanishing temperature as well as vanishing entropy., Comment: 30 pages, 5 pictures; V_2: a problem in the proof of the first law has been corrected. Results remain unchanged. Geometric interpretation and presentation improved; V_3: matches published version