-
Particle detectors under chronological hazard
Authors:
Ana Alonso-Serrano,
Erickson Tjoa,
Luis J. Garay,
Eduardo Martín-Martínez
Abstract:
We analyze how the presence of closed timelike curves (CTCs) characterizing a time machine can be discerned by placing a local particle detector in a region of spacetime which is causally disconnected from the CTCs. Our study shows that not only can the detector tell if there are CTCs, but also that the detector can separate topological from geometrical information and distinguish periodic spaceti…
▽ More
We analyze how the presence of closed timelike curves (CTCs) characterizing a time machine can be discerned by placing a local particle detector in a region of spacetime which is causally disconnected from the CTCs. Our study shows that not only can the detector tell if there are CTCs, but also that the detector can separate topological from geometrical information and distinguish periodic spacetimes without CTCs (like the Einstein cylinder), curvature, and spacetimes with topological identifications that enable time-machines.
△ Less
Submitted 27 February, 2024;
originally announced February 2024.
-
The Unruh-DeWitt model and its joint interacting Hilbert space
Authors:
Erickson Tjoa,
Finnian Gray
Abstract:
In this work we make the connection between the Unruh-DeWitt particle detector model applied to quantum field theory in curved spacetimes and the rigorous construction of the spin-boson model. With some modifications, we show that existing results about the existence of a spin-boson ground state can be adapted to the Unruh-DeWitt model. In the most relevant scenario involving massless scalar field…
▽ More
In this work we make the connection between the Unruh-DeWitt particle detector model applied to quantum field theory in curved spacetimes and the rigorous construction of the spin-boson model. With some modifications, we show that existing results about the existence of a spin-boson ground state can be adapted to the Unruh-DeWitt model. In the most relevant scenario involving massless scalar fields in (3+1)-dimensional globally hyperbolic spacetimes, where the Unruh-DeWitt model describes a simplified model of light-matter interaction, we argue that common choices of the spacetime smearing functions regulate the ultraviolet behaviour of the model but can still exhibit infrared divergences. In particular, this implies the well-known expectation that the joint interacting Hilbert space of the model cannot be described by the tensor product of a two-dimensional complex Hilbert space and the Fock space of the vacuum representation. We discuss the conditions under which this problem does not arise and the relevance of the operator-algebraic approach for better understanding of particle detector models and their applications.
△ Less
Submitted 8 February, 2024;
originally announced February 2024.
-
Unruh phenomena and thermalization for qudit detectors
Authors:
Caroline Lima,
Everett Patterson,
Erickson Tjoa,
Robert B. Mann
Abstract:
We study Unruh phenomena for a qudit detector coupled to a quantized scalar field, comparing its response to that of a standard qubit-based Unruh-DeWitt detector. We show that there are limitations to the utility of the detailed balance condition as an indicator for Unruh thermality of higher-dimensional qudit detector models. This can be traced to the fact that a qudit has multiple possible trans…
▽ More
We study Unruh phenomena for a qudit detector coupled to a quantized scalar field, comparing its response to that of a standard qubit-based Unruh-DeWitt detector. We show that there are limitations to the utility of the detailed balance condition as an indicator for Unruh thermality of higher-dimensional qudit detector models. This can be traced to the fact that a qudit has multiple possible transition channels between its energy levels, in contrast to the 2-level qubit model. We illustrate these limitations using two types of qutrit detector models based on the spin-1 representations of $SU(2)$ and the non-Hermitian generalization of the Pauli observables (the Heisenberg-Weyl operators).
△ Less
Submitted 12 February, 2024; v1 submitted 8 September, 2023;
originally announced September 2023.
-
Wigner function for quantum field theory via spacetime tiling
Authors:
Erickson Tjoa
Abstract:
We present a construction of the Wigner function for a bosonic quantum field theory that has well-defined ultraviolet (UV) and infrared (IR) properties. Our construction uses the local mode formalism in algebraic quantum field theory that is valid in any globally hyperbolic curved spacetimes, i.e., without invoking the path integral formalism. The idea is to build $N$ quantum harmonic oscillators…
▽ More
We present a construction of the Wigner function for a bosonic quantum field theory that has well-defined ultraviolet (UV) and infrared (IR) properties. Our construction uses the local mode formalism in algebraic quantum field theory that is valid in any globally hyperbolic curved spacetimes, i.e., without invoking the path integral formalism. The idea is to build $N$ quantum harmonic oscillators degrees of freedom from $2N$ smeared field operators and use them to "tile" a Cauchy surface of the spacetime manifold. The smallest support of the smearing functions that define each local mode define the UV scale and the number of modes local modes fix the IR scale. This construction can be viewed as a form of "covariant discretization" of the quantum field in curved spacetimes, since the tiling of the Cauchy surface does not depend on any choice of coordinate systems or foliation.
△ Less
Submitted 30 May, 2023;
originally announced May 2023.
-
Fuzzy spacetime: fundamental limits of quantum-optical holographic bulk reconstruction
Authors:
Erickson Tjoa
Abstract:
In this Essay we construct a concrete, non-perturbative realization of metric reconstruction using quantum-optical model of particle detectors in relativistic quantum information. The non-perturbative approach allows us to realize a version of "short-distance physics corresponds to poor statistics" idea by Kempf which occurs way above the Planck scale before one reaches the quantum-gravitational r…
▽ More
In this Essay we construct a concrete, non-perturbative realization of metric reconstruction using quantum-optical model of particle detectors in relativistic quantum information. The non-perturbative approach allows us to realize a version of "short-distance physics corresponds to poor statistics" idea by Kempf which occurs way above the Planck scale before one reaches the quantum-gravitational regime. In particular, the "fuzziness" of spacetime that arise from operational measurement protocols can be given a holographic dual interpretation using bulk-to-boundary correspondence between scalar correlators in asymptotically flat spacetimes. The holographic interpretation necessitates imperfect metric reconstruction even in principle due to the universality of future null infinity.
△ Less
Submitted 30 May, 2023; v1 submitted 28 March, 2023;
originally announced March 2023.
-
Effective master equations for two accelerated qubits
Authors:
Greg Kaplanek,
Erickson Tjoa
Abstract:
We revisit the problem involving two constantly accelerating Unruh-DeWitt detectors using Open Effective Field Theory methods. We study the time evolution of the joint detector state using a Markovian approximation which differs from the standard one taken in the literature. We show that this Markovian limit already implies the complete positivity of the dynamical evolution map without invoking th…
▽ More
We revisit the problem involving two constantly accelerating Unruh-DeWitt detectors using Open Effective Field Theory methods. We study the time evolution of the joint detector state using a Markovian approximation which differs from the standard one taken in the literature. We show that this Markovian limit already implies the complete positivity of the dynamical evolution map without invoking the rotating wave approximation (RWA), in contrast to standard derivations of open system master equations. By calculating explicitly the domain of validity of this Markovian approximation, we argue that the lack of complete positivity in the usual microscopic derivation stems from the (subtle) fact that the Redfield equation is used outside its domain of validity. We give two well-known cases studied in the literature that violate the validity of the Markovian approximation: (i) the ``stacked trajectory'' limit (when detector trajectories are taken to be on top of one another), and (ii) large gap-to-acceleration ratio. Since Markovian dynamics with or without RWA can lead to different qualitative predictions for entanglement dynamics, our work emphasizes the need to properly track the regime of validity of all approximations.
△ Less
Submitted 24 January, 2023; v1 submitted 27 July, 2022;
originally announced July 2022.
-
Non-perturbative simple-generated interactions with a quantum field for arbitrary Gaussian states
Authors:
Erickson Tjoa
Abstract:
In this work we first collect and generalize several existing non-perturbative models for the interaction between a single two-level qubit detector and a relativistic quantum scalar field in arbitrary curved spacetimes, where the time evolution is given by simple-generated unitaries, i.e., those generated by Schmidt rank-1 interaction Hamiltonians. We then extend the relativistic quantum channel a…
▽ More
In this work we first collect and generalize several existing non-perturbative models for the interaction between a single two-level qubit detector and a relativistic quantum scalar field in arbitrary curved spacetimes, where the time evolution is given by simple-generated unitaries, i.e., those generated by Schmidt rank-1 interaction Hamiltonians. We then extend the relativistic quantum channel associated to these non-perturbative models to include a very large class of Gaussian states of the quantum field, that includes an arbitrary combinations of coherent and squeezing operations (i.e., Gaussian operations) on the field. We show that all physical results involving the non-vacuum Gaussian states can be rephrased in terms of interaction with the vacuum state but with Gaussian operators applied to the field operators via the adjoint channel, effectively giving a "Fourier transformed" interpretation of the Gaussian operations in terms of the causal propagators in spacetime. Furthermore, we show that in these non-perturbative models it is possible to perform exact computation of the Rényi entropy and hence, via the replica trick, the von Neumann entropy for the field state after the interaction with the detector, without making any assumptions about the purity of the joint initial states of the detector and the field. This gives us a three-parameter family of "generalized cat states" of the field whose entropies are finite and exactly computable.
△ Less
Submitted 14 April, 2023; v1 submitted 3 July, 2022;
originally announced July 2022.
-
Quantum teleportation with relativistic communication from first principles
Authors:
Erickson Tjoa
Abstract:
In this work we provide a genuine relativistic quantum teleportation protocol whose classical communication component makes use of relativistic causal propagation of a quantum field. Consequently, the quantum teleportation is fully relativistic by construction. Our scheme is based on Unruh-DeWitt qubit detector model, where the quantum state being teleported is associated to an actual qubit rather…
▽ More
In this work we provide a genuine relativistic quantum teleportation protocol whose classical communication component makes use of relativistic causal propagation of a quantum field. Consequently, the quantum teleportation is fully relativistic by construction. Our scheme is based on Unruh-DeWitt qubit detector model, where the quantum state being teleported is associated to an actual qubit rather than a field mode considered by Alsing and Milburn [PRL 91, 180404 (2003)]. We show that the existing works in (relativistic) quantum information, including good definitions of one-shot and asymptotic channel capacities, as well as algebraic formulation of quantum field theory, already provide us with all the necessary ingredients to construct fundamentally relativistic teleportation protocol in relatively straightforward manner.
△ Less
Submitted 29 September, 2022; v1 submitted 18 June, 2022;
originally announced June 2022.
-
Fermi two-atom problem: non-perturbative approach via relativistic quantum information and algebraic quantum field theory
Authors:
Erickson Tjoa
Abstract:
In this work we revisit the famous Fermi two-atom problem, which concerns how relativistic causality impacts atomic transition probabilities, using the tools from relativistic quantum information (RQI) and algebraic quantum field theory (AQFT). The problem has sparked different analyses from many directions and angles since the proposed solution by Buchholz and Yngvason (1994). Some of these analy…
▽ More
In this work we revisit the famous Fermi two-atom problem, which concerns how relativistic causality impacts atomic transition probabilities, using the tools from relativistic quantum information (RQI) and algebraic quantum field theory (AQFT). The problem has sparked different analyses from many directions and angles since the proposed solution by Buchholz and Yngvason (1994). Some of these analyses employ various approximations, heuristics, perturbative methods, which tends to render some of the otherwise useful insights somewhat obscured. It is also noted that they are all studied in flat spacetime. We show that current tools in relativistic quantum information, combined with algebraic approach to quantum field theory, are now powerful enough to provide fuller and cleaner analysis of the Fermi two-atom problem for arbitrary curved spacetimes in a completely non-perturbative manner. Our result gives the original solution of Buchholz and Yngvason a very operational reinterpretation in terms of qubits interacting with a quantum field, and allows for various natural generalizations and inclusion of detector-based local measurement for the quantum field (Phys. Rev. D 105, 065003).
△ Less
Submitted 16 August, 2022; v1 submitted 5 June, 2022;
originally announced June 2022.
-
Modest holography and bulk reconstruction in asymptotically flat spacetimes
Authors:
Erickson Tjoa,
Finnian Gray
Abstract:
In this work we present a "modest" holographic reconstruction of the bulk geometry in asymptotically flat spacetime using the two-point correlators of boundary quantum field theory (QFT) in asymptotically flat spacetime. The boundary QFT lives on the null boundary of the spacetime, namely null infinity and/or the Killing horizons. The bulk reconstruction relies on two unrelated results: (i) there…
▽ More
In this work we present a "modest" holographic reconstruction of the bulk geometry in asymptotically flat spacetime using the two-point correlators of boundary quantum field theory (QFT) in asymptotically flat spacetime. The boundary QFT lives on the null boundary of the spacetime, namely null infinity and/or the Killing horizons. The bulk reconstruction relies on two unrelated results: (i) there is a bulk-to-boundary type correspondence between free quantum fields living in the bulk manifold and free quantum fields living on its null boundary, and (ii) one can construct the metric by making use of the Hadamard expansion of the field living in the bulk. This holographic reconstruction is "modest" in that the fields used are non-interacting and not strong-weak holographic duality in the sense of AdS/CFT, but it works for generic asymptotically flat spacetime subject to some reasonably mild conditions.
△ Less
Submitted 2 August, 2022; v1 submitted 27 April, 2022;
originally announced April 2022.
-
Holographic reconstruction of asymptotically flat spacetimes
Authors:
Erickson Tjoa,
Finnian Gray
Abstract:
We present a "holographic" reconstruction of bulk spacetime geometry using correlation functions of a massless field living at the "future boundary" of the spacetime, namely future null infinity $\mathscr{I}^+$. It is holographic in the sense that there exists a one-to-one correspondence between correlation functions of a massless field in four-dimensional spacetime $\mathcal{M}$ and those of anot…
▽ More
We present a "holographic" reconstruction of bulk spacetime geometry using correlation functions of a massless field living at the "future boundary" of the spacetime, namely future null infinity $\mathscr{I}^+$. It is holographic in the sense that there exists a one-to-one correspondence between correlation functions of a massless field in four-dimensional spacetime $\mathcal{M}$ and those of another massless field living in three-dimensional null boundary $\mathscr{I}^+$. The idea is to first reconstruct the bulk metric $g_{μν}$ by "inverting" the bulk correlation functions and re-express the latter in terms of boundary correlators via the correspondence. This effectively allows asymptotic observers close to $\mathscr{I}^+$ to reconstruct the deep interior of the spacetime using only correlation functions localized near $\mathscr{I}^+$.
△ Less
Submitted 23 May, 2022; v1 submitted 29 March, 2022;
originally announced March 2022.
-
Channel capacity of relativistic quantum communication with rapid interaction
Authors:
Erickson Tjoa,
Kensuke Gallock-Yoshimura
Abstract:
In this work we study nonperturbatively the transmission of classical and quantum information in globally hyperbolic spacetimes, where the communication channel is between two qubit detectors interacting with a quantized massless scalar field via delta-coupling interaction. This interaction approximates very rapid detector-field interaction, effectively occurring at a single instant in time for ea…
▽ More
In this work we study nonperturbatively the transmission of classical and quantum information in globally hyperbolic spacetimes, where the communication channel is between two qubit detectors interacting with a quantized massless scalar field via delta-coupling interaction. This interaction approximates very rapid detector-field interaction, effectively occurring at a single instant in time for each detector. We show that when both detectors interact via delta-coupling, one can arrange and tune the detectors so that the channel capacity is (at least) as good as the quantum channel constructed nonperturbatively using \textit{gapless detectors} by Landulfo [PRD 93, 104019]. Furthermore, we prove that this channel capacity is in fact optimal, i.e., both nonperturbative methods give essentially the same channel capacity, thus there is a sense in which the two methods can be regarded as equivalent as far as relativistic quantum communication is concerned.
△ Less
Submitted 30 May, 2023; v1 submitted 24 February, 2022;
originally announced February 2022.
-
Unruh-DeWitt detector in dimensionally-reduced static spherically symmetric spacetimes
Authors:
Erickson Tjoa,
Robert B. Mann
Abstract:
We study the dynamics of an Unruh-DeWitt detector interacting with a massless scalar field in an arbitrary static spherically symmetric spacetimes whose metric is characterised by a single metric function $f(r)$. In order to obtain clean physical insights, we employ the derivative coupling variant of the Unruh-DeWitt model in (1+1) dimensions where powerful conformal techniques enable closed-form…
▽ More
We study the dynamics of an Unruh-DeWitt detector interacting with a massless scalar field in an arbitrary static spherically symmetric spacetimes whose metric is characterised by a single metric function $f(r)$. In order to obtain clean physical insights, we employ the derivative coupling variant of the Unruh-DeWitt model in (1+1) dimensions where powerful conformal techniques enable closed-form expressions for the vacuum two-point functions. Due to the generality of the formalism, we will be able to study a very general class of static spherically symmetric (SSS) background. We pick three examples to illustrate our method: (1) non-singular Hayward black holes, (2) the recently discovered $D\to 4$ limit of Gauss-Bonnet black holes, and (3) the "black bounce" metric that interpolates Schwarzschild black holes and traversable wormholes. We also show that the derivative coupling Wightman function associated with the generalized Hartle-Hawking vacuum satisfies the KMS property with the well-known temperature $f'(r_\text{H})/(4π)$, where $r_\text{H}$ is the horizon radius.
△ Less
Submitted 4 March, 2022; v1 submitted 8 February, 2022;
originally announced February 2022.
-
Improving Deep Neural Network Classification Confidence using Heatmap-based eXplainable AI
Authors:
Erico Tjoa,
Hong **g Khok,
Tushar Chouhan,
Guan Cuntai
Abstract:
This paper quantifies the quality of heatmap-based eXplainable AI (XAI) methods w.r.t image classification problem. Here, a heatmap is considered desirable if it improves the probability of predicting the correct classes. Different XAI heatmap-based methods are empirically shown to improve classification confidence to different extents depending on the datasets, e.g. Saliency works best on ImageNe…
▽ More
This paper quantifies the quality of heatmap-based eXplainable AI (XAI) methods w.r.t image classification problem. Here, a heatmap is considered desirable if it improves the probability of predicting the correct classes. Different XAI heatmap-based methods are empirically shown to improve classification confidence to different extents depending on the datasets, e.g. Saliency works best on ImageNet and Deconvolution on Chest X-Ray Pneumonia dataset. The novelty includes a new gap distribution that shows a stark difference between correct and wrong predictions. Finally, the generative augmentative explanation is introduced, a method to generate heatmaps capable of improving predictive confidence to a high level.
△ Less
Submitted 20 January, 2023; v1 submitted 30 December, 2021;
originally announced January 2022.
-
Self Reward Design with Fine-grained Interpretability
Authors:
Erico Tjoa,
Guan Cuntai
Abstract:
The black-box nature of deep neural networks (DNN) has brought to attention the issues of transparency and fairness. Deep Reinforcement Learning (Deep RL or DRL), which uses DNN to learn its policy, value functions etc, is thus also subject to similar concerns. This paper proposes a way to circumvent the issues through the bottom-up design of neural networks with detailed interpretability, where e…
▽ More
The black-box nature of deep neural networks (DNN) has brought to attention the issues of transparency and fairness. Deep Reinforcement Learning (Deep RL or DRL), which uses DNN to learn its policy, value functions etc, is thus also subject to similar concerns. This paper proposes a way to circumvent the issues through the bottom-up design of neural networks with detailed interpretability, where each neuron or layer has its own meaning and utility that corresponds to humanly understandable concept. The framework introduced in this paper is called the Self Reward Design (SRD), inspired by the Inverse Reward Design, and this interpretable design can (1) solve the problem by pure design (although imperfectly) and (2) be optimized like a standard DNN. With deliberate human designs, we show that some RL problems such as lavaland and MuJoCo can be solved using a model constructed with standard NN components with few parameters. Furthermore, with our fish sale auction example, we demonstrate how SRD is used to address situations that will not make sense if black-box models are used, where humanly-understandable semantic-based decision is required.
△ Less
Submitted 20 January, 2023; v1 submitted 30 December, 2021;
originally announced December 2021.
-
Two Instances of Interpretable Neural Network for Universal Approximations
Authors:
Erico Tjoa,
Guan Cuntai
Abstract:
This paper proposes two bottom-up interpretable neural network (NN) constructions for universal approximation, namely Triangularly-constructed NN (TNN) and Semi-Quantized Activation NN (SQANN). Further notable properties are (1) resistance to catastrophic forgetting (2) existence of proof for arbitrarily high accuracies (3) the ability to identify samples that are out-of-distribution through inter…
▽ More
This paper proposes two bottom-up interpretable neural network (NN) constructions for universal approximation, namely Triangularly-constructed NN (TNN) and Semi-Quantized Activation NN (SQANN). Further notable properties are (1) resistance to catastrophic forgetting (2) existence of proof for arbitrarily high accuracies (3) the ability to identify samples that are out-of-distribution through interpretable activation "fingerprints".
△ Less
Submitted 9 May, 2022; v1 submitted 30 December, 2021;
originally announced December 2021.
-
When entanglement harvesting is not really harvesting
Authors:
Erickson Tjoa,
Eduardo Martín-Martínez
Abstract:
We revisit the entanglement harvesting protocol when two detectors are in causal contact. We study the role of field-mediated communication in generating entanglement between the two detectors interacting with a quantum field. We provide a quantitative estimator of the relative contribution of communication versus genuine entanglement harvesting. For massless scalar fields in flat spacetime, we sh…
▽ More
We revisit the entanglement harvesting protocol when two detectors are in causal contact. We study the role of field-mediated communication in generating entanglement between the two detectors interacting with a quantum field. We provide a quantitative estimator of the relative contribution of communication versus genuine entanglement harvesting. For massless scalar fields in flat spacetime, we show that when two detectors can communicate via the field, the detectors do not really harvest entanglement from the field, and instead they get entangled only via the field-mediated communication channel. In other words, in these scenarios the entanglement harvesting protocol is truly "harvesting entanglement" from the field only when the detectors are not able to communicate. In contrast, for massive scalar fields both communication and genuine harvesting contribute equally to the bipartite entanglement when the detectors are causally connected. These results emphasize the importance of taking into account the causal relationships between two parties involved in this relativistic quantum information protocol before we can declare that it is truly entanglement harvesting.
△ Less
Submitted 30 May, 2023; v1 submitted 23 September, 2021;
originally announced September 2021.
-
The time traveler's guide to the quantization of zero modes
Authors:
Ana Alonso-Serrano,
Erickson Tjoa,
Luis J. Garay,
Eduardo Martín-Martínez
Abstract:
We study the relationship between the quantization of a massless scalar field on the two-dimensional Einstein cylinder and in a spacetime with a time machine. We find that the latter picks out a unique prescription for the state of the zero mode in the Einstein cylinder. We show how this choice arises from the computation of the vacuum Wightman function and the vacuum renormalized stress-energy te…
▽ More
We study the relationship between the quantization of a massless scalar field on the two-dimensional Einstein cylinder and in a spacetime with a time machine. We find that the latter picks out a unique prescription for the state of the zero mode in the Einstein cylinder. We show how this choice arises from the computation of the vacuum Wightman function and the vacuum renormalized stress-energy tensor in the time-machine geometry. Finally, we relate the previously proposed regularization of the zero mode state as a squeezed state with the time-machine warp parameter, thus demonstrating that the quantization in the latter regularizes the quantization in an Einstein cylinder.
△ Less
Submitted 26 December, 2021; v1 submitted 16 August, 2021;
originally announced August 2021.
-
Quantum imprints of gravitational shockwaves
Authors:
Finnian Gray,
David Kubiznak,
Taillte May,
Sydney Timmerman,
Erickson Tjoa
Abstract:
Gravitational shockwaves are simple exact solutions of Einstein equations representing the fields of ultrarelativistic sources and idealized gravitational waves (shocks). Historically, much work has focused on shockwaves in the context of possible black hole formation in high energy particle collisions, yet they remain at the forefront of research even today. Representing hard modes in the bulk, s…
▽ More
Gravitational shockwaves are simple exact solutions of Einstein equations representing the fields of ultrarelativistic sources and idealized gravitational waves (shocks). Historically, much work has focused on shockwaves in the context of possible black hole formation in high energy particle collisions, yet they remain at the forefront of research even today. Representing hard modes in the bulk, shocks give rise to the gravitational memory effect at the classical level and implant supertranslation (BMS) hair onto a classical spacetime at the quantum level. The aim of this paper is to further our understanding of the `information content' of such supertranslations. Namely, we show that, contrary to the several claims in the literature, a gravitational shockwave does leave a quantum imprint on the vacuum state of a test quantum field and that this imprint is accessible to local observers carrying Unruh--DeWitt (UDW) detectors in this spacetime.
△ Less
Submitted 3 June, 2021; v1 submitted 19 May, 2021;
originally announced May 2021.
-
A Modified Convolutional Network for Auto-encoding based on Pattern Theory Growth Function
Authors:
Erico Tjoa
Abstract:
This brief paper reports the shortcoming of a variant of convolutional neural network whose components are developed based on the pattern theory framework.
This brief paper reports the shortcoming of a variant of convolutional neural network whose components are developed based on the pattern theory framework.
△ Less
Submitted 4 April, 2021;
originally announced April 2021.
-
Harvesting Entanglement with Detectors Freely Falling into a Black Hole
Authors:
Kensuke Gallock-Yoshimura,
Erickson Tjoa,
Robert B. Mann
Abstract:
We carry out the first investigation of the entanglement and mutual information harvesting protocols for detectors freely falling into a black hole. Working in $(1+1)$-dimensional Schwarzschild black hole spacetime, we consider two pointlike Unruh-DeWitt (UDW) detectors in different combinations of free-falling and static trajectories. Employing a generalization of relative velocity suitable for c…
▽ More
We carry out the first investigation of the entanglement and mutual information harvesting protocols for detectors freely falling into a black hole. Working in $(1+1)$-dimensional Schwarzschild black hole spacetime, we consider two pointlike Unruh-DeWitt (UDW) detectors in different combinations of free-falling and static trajectories. Employing a generalization of relative velocity suitable for curved spacetimes, we find that the amount of correlations extracted from the black hole vacuum, at least outside the near-horizon regime, is largely kinematic in origin (i.e. it is mostly due to the relative velocities of the detectors). Second, correlations can be harvested purely from the black hole vacuum even when the detectors are causally disconnected by the event horizon. Finally, we show that the previously known `entanglement shadow' near the horizon is indeed absent for the case of two free-falling-detectors, since their relative gravitational redshift remains finite as the horizon is crossed, in accordance with the equivalence principle.
△ Less
Submitted 5 July, 2021; v1 submitted 18 February, 2021;
originally announced February 2021.
-
What makes a particle detector click
Authors:
Erickson Tjoa,
Irene López Gutiérrez,
Allison Sachs,
Eduardo Martín-Martínez
Abstract:
We highlight fundamental differences in the models of light-matter interaction between the behaviour of Fock state detection in free space versus optical cavities. To do so, we study the phenomenon of resonance of detectors with Fock wavepackets as a function of their degree of monochromaticity, the number of spatial dimensions, the linear or quadratic nature of the light-matter coupling, and the…
▽ More
We highlight fundamental differences in the models of light-matter interaction between the behaviour of Fock state detection in free space versus optical cavities. To do so, we study the phenomenon of resonance of detectors with Fock wavepackets as a function of their degree of monochromaticity, the number of spatial dimensions, the linear or quadratic nature of the light-matter coupling, and the presence (or absence) of cavity walls in space. In doing so we show that intuition coming from quantum optics in cavities does not straightforwardly carry to the free space case. For example, in $(3+1)$ dimensions the detector response to a Fock wavepacket will go to zero as the wavepacket is made more and more monochromatic and in coincidence with the detector's resonant frequency. This is so even though the energy of the free-space wavepacket goes to the expected finite value of $\hbarΩ$ in the monochromatic limit. This is in contrast to the behaviour of the light-matter interaction in a cavity (even a large one) where the probability of absorbing a Fock quantum is maximized when the quantum is more monochromatic at the detector's resonance frequency. We trace this crucial difference to the fact that monochromatic Fock states are not normalizable in the continuum, thus physical Fock states need to be constructed out of normalizable wavepackets whose energy density goes to zero in the monochromatic limit as they get spatially delocalized.
△ Less
Submitted 6 July, 2021; v1 submitted 10 February, 2021;
originally announced February 2021.
-
Convolutional Neural Network Interpretability with General Pattern Theory
Authors:
Erico Tjoa,
Guan Cuntai
Abstract:
Ongoing efforts to understand deep neural networks (DNN) have provided many insights, but DNNs remain incompletely understood. Improving DNN's interpretability has practical benefits, such as more accountable usage, better algorithm maintenance and improvement. The complexity of dataset structure may contribute to the difficulty in solving interpretability problem arising from DNN's black-box mech…
▽ More
Ongoing efforts to understand deep neural networks (DNN) have provided many insights, but DNNs remain incompletely understood. Improving DNN's interpretability has practical benefits, such as more accountable usage, better algorithm maintenance and improvement. The complexity of dataset structure may contribute to the difficulty in solving interpretability problem arising from DNN's black-box mechanism. Thus, we propose to use pattern theory formulated by Ulf Grenander, in which data can be described as configurations of fundamental objects that allow us to investigate convolutional neural network's (CNN) interpretability in a component-wise manner. Specifically, U-Net-like structure is formed by attaching expansion blocks (EB) to ResNet, allowing it to perform semantic segmentation-like tasks at its EB output channels designed to be compatible with pattern theory's configurations. Through these modules, some heatmap-based explainable artificial intelligence (XAI) methods will be shown to extract explanations w.r.t individual generators that make up a single data sample, potentially reducing the impact of dataset's complexity to interpretability problem. The MNIST-equivalent dataset containing pattern theory's elements is designed to facilitate smoother entry into this framework, along which the theory's generative aspect is naturally presented.
△ Less
Submitted 5 February, 2021;
originally announced February 2021.
-
Quantifying Explainability of Saliency Methods in Deep Neural Networks with a Synthetic Dataset
Authors:
Erico Tjoa,
Cuntai Guan
Abstract:
Post-hoc analysis is a popular category in eXplainable artificial intelligence (XAI) study. In particular, methods that generate heatmaps have been used to explain the deep neural network (DNN), a black-box model. Heatmaps can be appealing due to the intuitive and visual ways to understand them but assessing their qualities might not be straightforward. Different ways to assess heatmaps' quality h…
▽ More
Post-hoc analysis is a popular category in eXplainable artificial intelligence (XAI) study. In particular, methods that generate heatmaps have been used to explain the deep neural network (DNN), a black-box model. Heatmaps can be appealing due to the intuitive and visual ways to understand them but assessing their qualities might not be straightforward. Different ways to assess heatmaps' quality have their own merits and shortcomings. This paper introduces a synthetic dataset that can be generated adhoc along with the ground-truth heatmaps for more objective quantitative assessment. Each sample data is an image of a cell with easily recognized features that are distinguished from localization ground-truth mask, hence facilitating a more transparent assessment of different XAI methods. Comparison and recommendations are made, shortcomings are clarified along with suggestions for future research directions to handle the finer details of select post-hoc analysis methods. Furthermore, mabCAM is introduced as the heatmap generation method compatible with our ground-truth heatmaps. The framework is easily generalizable and uses only standard deep learning components.
△ Less
Submitted 10 December, 2022; v1 submitted 7 September, 2020;
originally announced September 2020.
-
Generalization on the Enhancement of Layerwise Relevance Interpretability of Deep Neural Network
Authors:
Erico Tjoa,
Guan Cuntai
Abstract:
The practical application of deep neural networks are still limited by their lack of transparency. One of the efforts to provide explanation for decisions made by artificial intelligence (AI) is the use of saliency or heat maps highlighting relevant regions that contribute significantly to its prediction. A layer-wise amplitude filtering method was previously introduced to improve the quality of h…
▽ More
The practical application of deep neural networks are still limited by their lack of transparency. One of the efforts to provide explanation for decisions made by artificial intelligence (AI) is the use of saliency or heat maps highlighting relevant regions that contribute significantly to its prediction. A layer-wise amplitude filtering method was previously introduced to improve the quality of heatmaps, performing error corrections by noise-spike suppression. In this study, we generalize the layerwise error correction by considering any identifiable error and assuming there exists a groundtruth interpretable information. The forms of errors propagated through layerwise relevance methods are studied and we propose a filtering technique for interpretability signal rectification taylored to the trend of signal amplitude of the particular neural network used. Finally, we put forth arguments for the use of groundtruth interpretable information.
△ Less
Submitted 18 October, 2020; v1 submitted 5 September, 2020;
originally announced September 2020.
-
Harvesting correlations in Schwarzschild and collapsing shell spacetimes
Authors:
Erickson Tjoa,
Robert B. Mann
Abstract:
We study the harvesting of correlations by two Unruh-DeWitt static detectors from the vacuum state of a massless scalar field in a background Vaidya spacetime consisting of a collapsing null shell that forms a Schwarzschild black hole (hereafter Vaidya spacetime for brevity), and we compare the results with those associated with the three preferred vacua (Boulware, Unruh, Hartle-Hawking-Israel vac…
▽ More
We study the harvesting of correlations by two Unruh-DeWitt static detectors from the vacuum state of a massless scalar field in a background Vaidya spacetime consisting of a collapsing null shell that forms a Schwarzschild black hole (hereafter Vaidya spacetime for brevity), and we compare the results with those associated with the three preferred vacua (Boulware, Unruh, Hartle-Hawking-Israel vacua) of the eternal Schwarzschild black hole spacetime. To do this we make use of the explicit Wightman functions for a massless scalar field available in (1+1)-dimensional models of the collapsing spacetime and Schwarzschild spacetimes, and the detectors couple to the proper time derivative of the field. First we find that, with respect to the harvesting protocol, the Unruh vacuum agrees very well with the Vaidya vacuum near the horizon even for finite-time interactions. Second, all four vacua have different capacities for creating correlations between the detectors, with the Vaidya vacuum interpolating between the Unruh vacuum near the horizon and the Boulware vacuum far from the horizon. Third, we show that the black hole horizon inhibits \textit{any} correlations, not just entanglement. Finally, we show that the efficiency of the harvesting protocol depend strongly on the signalling ability of the detectors, which is highly non-trivial in presence of curvature. We provide an asymptotic analysis of the Vaidya vacuum to clarify the relationship between the Boulware/Unruh interpolation and the near/far from horizon and early/late-time limits. We demonstrate a straightforward implementation of numerical contour integration to perform all the calculations.
△ Less
Submitted 7 January, 2021; v1 submitted 6 July, 2020;
originally announced July 2020.
-
Vacuum entanglement harvesting with a zero mode
Authors:
Erickson Tjoa,
Eduardo Martín-Martínez
Abstract:
We investigate vacuum entanglement harvesting in the presence of a zero mode. We show that, for a variety of detector models and couplings (namely, Unruh-DeWitt qubit and harmonic oscillator detectors, amplitude and derivative coupling), the results are strongly dependent on the state of the zero mode, revealing an ambiguity in studies of entanglement harvesting with Neumann or periodic boundary c…
▽ More
We investigate vacuum entanglement harvesting in the presence of a zero mode. We show that, for a variety of detector models and couplings (namely, Unruh-DeWitt qubit and harmonic oscillator detectors, amplitude and derivative coupling), the results are strongly dependent on the state of the zero mode, revealing an ambiguity in studies of entanglement harvesting with Neumann or periodic boundary conditions, or in general in spacetimes with toroidal topologies.
△ Less
Submitted 6 July, 2021; v1 submitted 26 February, 2020;
originally announced February 2020.
-
Enhancing the Extraction of Interpretable Information for Ischemic Stroke Imaging from Deep Neural Networks
Authors:
Erico Tjoa,
Guo Heng,
Lu Yuhao,
Cuntai Guan
Abstract:
We implement a visual interpretability method Layer-wise Relevance Propagation (LRP) on top of 3D U-Net trained to perform lesion segmentation on the small dataset of multi-modal images provided by ISLES 2017 competition. We demonstrate that LRP modifications could provide more sensible visual explanations to an otherwise highly noise-skewed saliency map. We also link amplitude of modified signals…
▽ More
We implement a visual interpretability method Layer-wise Relevance Propagation (LRP) on top of 3D U-Net trained to perform lesion segmentation on the small dataset of multi-modal images provided by ISLES 2017 competition. We demonstrate that LRP modifications could provide more sensible visual explanations to an otherwise highly noise-skewed saliency map. We also link amplitude of modified signals to useful information content. High amplitude localized signals appear to constitute the noise that undermines the interpretability capacity of LRP. Furthermore, mathematical framework for possible analysis of function approximation is developed by analogy.
△ Less
Submitted 13 January, 2020; v1 submitted 19 November, 2019;
originally announced November 2019.
-
A Survey on Explainable Artificial Intelligence (XAI): Towards Medical XAI
Authors:
Erico Tjoa,
Cuntai Guan
Abstract:
Recently, artificial intelligence and machine learning in general have demonstrated remarkable performances in many tasks, from image processing to natural language processing, especially with the advent of deep learning. Along with research progress, they have encroached upon many different fields and disciplines. Some of them require high level of accountability and thus transparency, for exampl…
▽ More
Recently, artificial intelligence and machine learning in general have demonstrated remarkable performances in many tasks, from image processing to natural language processing, especially with the advent of deep learning. Along with research progress, they have encroached upon many different fields and disciplines. Some of them require high level of accountability and thus transparency, for example the medical sector. Explanations for machine decisions and predictions are thus needed to justify their reliability. This requires greater interpretability, which often means we need to understand the mechanism underlying the algorithms. Unfortunately, the blackbox nature of the deep learning is still unresolved, and many machine decisions are still poorly understood. We provide a review on interpretabilities suggested by different research works and categorize them. The different categories show different dimensions in interpretability research, from approaches that provide "obviously" interpretable information to the studies of complex patterns. By applying the same categorization to interpretability in medical research, it is hoped that (1) clinicians and practitioners can subsequently approach these methods with caution, (2) insights into interpretability will be born with more considerations for medical practices, and (3) initiatives to push forward data-based, mathematically- and technically-grounded medical education are encouraged.
△ Less
Submitted 10 August, 2020; v1 submitted 17 July, 2019;
originally announced July 2019.
-
Zero mode suppression of superluminal signals in light-matter interactions
Authors:
Erickson Tjoa,
Eduardo Martin-Martinez
Abstract:
We show how two Unruh-DeWitt detectors that do not couple to the zero mode of a quantum field can exchange information faster than the speed of light. We analyze the specific cases of periodic and Neumann boundary conditions in flat spacetime with arbitrary spatial dimensions, and we show that the superluminal signal strength is only polynomially suppressed with the distance to the lightcone. Ther…
▽ More
We show how two Unruh-DeWitt detectors that do not couple to the zero mode of a quantum field can exchange information faster than the speed of light. We analyze the specific cases of periodic and Neumann boundary conditions in flat spacetime with arbitrary spatial dimensions, and we show that the superluminal signal strength is only polynomially suppressed with the distance to the lightcone. Therefore, in any relativistic scenario modelling the light-matter interaction where a zero mode is present, particle detectors should explicitly couple to the zero mode.
△ Less
Submitted 26 March, 2019; v1 submitted 5 November, 2018;
originally announced November 2018.
-
Entanglement Harvesting with Moving Mirrors
Authors:
Wan Cong,
Erickson Tjoa,
Robert B. Mann
Abstract:
We study the phenomenon of entanglement extraction from the vacuum of a massless scalar field in $(1+1)$ dimensional spacetime in presence of a moving Dirichlet boundary condition, i.e. mirror spacetime, using two inertial Unruh-DeWitt detectors. We consider a variety of non-trivial trajectories for these accelerating mirrors and find (1) an entanglement inhibition phenomenon similar to that recen…
▽ More
We study the phenomenon of entanglement extraction from the vacuum of a massless scalar field in $(1+1)$ dimensional spacetime in presence of a moving Dirichlet boundary condition, i.e. mirror spacetime, using two inertial Unruh-DeWitt detectors. We consider a variety of non-trivial trajectories for these accelerating mirrors and find (1) an entanglement inhibition phenomenon similar to that recently seen for black holes, as well as (2) trajectory-independent entanglement enhancement in some regimes. We show that the qualitative result obtained is the same for both linear and derivative couplings of the detector with the field.
△ Less
Submitted 9 July, 2019; v1 submitted 16 October, 2018;
originally announced October 2018.
-
Particle Detectors, Cavities, and the Weak Equivalence Principle
Authors:
Erickson Tjoa,
Robert B. Mann,
Eduardo Martin-Martinez
Abstract:
We analyze a quantum version of the weak equivalence principle, in which we compare the response of a static particle detector crossed by an accelerated cavity with the response of an accelerated detector crossing a static cavity in (1+1)-dimensional flat spacetime. We show, for both massive and massless scalar fields, that the non-locality of the field is enough for the detector to distinguish th…
▽ More
We analyze a quantum version of the weak equivalence principle, in which we compare the response of a static particle detector crossed by an accelerated cavity with the response of an accelerated detector crossing a static cavity in (1+1)-dimensional flat spacetime. We show, for both massive and massless scalar fields, that the non-locality of the field is enough for the detector to distinguish the two scenarios. We find this result holds for vacuum and excited field states of different kinds and we clarify the role of field mass in this setup.
△ Less
Submitted 13 October, 2018; v1 submitted 19 July, 2018;
originally announced July 2018.
-
Thermodynamics of hairy black holes in Lovelock gravity
Authors:
Robie A. Hennigar,
Erickson Tjoa,
Robert B. Mann
Abstract:
We perform a thorough study of the thermodynamic properties of a class of Lovelock black holes with conformal scalar hair arising from coupling of a real scalar field to the dimensionally extended Euler densities. We study the linearized equations of motion of the theory and describe constraints under which the theory is free from ghosts/tachyons. We then consider, within the context of black hole…
▽ More
We perform a thorough study of the thermodynamic properties of a class of Lovelock black holes with conformal scalar hair arising from coupling of a real scalar field to the dimensionally extended Euler densities. We study the linearized equations of motion of the theory and describe constraints under which the theory is free from ghosts/tachyons. We then consider, within the context of black hole chemistry, the thermodynamics of the hairy black holes in the Gauss-Bonnet and cubic Lovelock theories. We clarify the connection between isolated critical points and thermodynamic singularities, finding a one parameter family of these critical points which occur for well-defined thermodynamic parameters. We also report on a number of novel results, including `virtual triple points' and the first example of a `$λ$-line'---a line of second order phase transitions---in black hole thermodynamics.
△ Less
Submitted 20 February, 2017; v1 submitted 20 December, 2016;
originally announced December 2016.
-
Superfluid Black Holes
Authors:
Robie A. Hennigar,
Robert B. Mann,
Erickson Tjoa
Abstract:
We present what we believe is the first example of a "$λ$-line" phase transition in black hole thermodynamics. This is a line of (continuous) second order phase transitions which in the case of liquid $^4$He marks the onset of superfluidity. The phase transition occurs for a class of asymptotically AdS hairy black holes in Lovelock gravity where a real scalar field is conformally coupled to gravit…
▽ More
We present what we believe is the first example of a "$λ$-line" phase transition in black hole thermodynamics. This is a line of (continuous) second order phase transitions which in the case of liquid $^4$He marks the onset of superfluidity. The phase transition occurs for a class of asymptotically AdS hairy black holes in Lovelock gravity where a real scalar field is conformally coupled to gravity. We discuss the origin of this phase transition and outline the circumstances under which it (or generalizations of it) could occur.
△ Less
Submitted 8 September, 2016;
originally announced September 2016.
