Search

We present the results of the "Fast Calorimeter Simulation Challenge 2022" - the CaloChallenge. We study state-of-the-art generative models on four calorimeter shower datasets of increasing dimensionality, ranging from a few hundred voxels to a few tens of thousand voxels. The 31 individual submissions span a wide range of current popular generative architectures, including Variational AutoEncoders (VAEs), Generative Adversarial Networks (GANs), Normalizing Flows, Diffusion models, and models based on Conditional Flow Matching. We compare all submissions in terms of quality of generated calorimeter showers, as well as shower generation time and model size. To assess the quality we use a broad range of different metrics including differences in 1-dimensional histograms of observables, KPD/FPD scores, AUCs of binary classifiers, and the log-posterior of a multiclass classifier. The results of the CaloChallenge provide the most complete and comprehensive survey of cutting-edge approaches to calorimeter fast simulation to date. In addition, our work provides a uniquely detailed perspective on the important problem of how to evaluate generative models. As such, the results presented here should be applicable for other domains that use generative AI and require fast and faithful generation of samples in a large phase space., Comment: 204 pages, 100+ figures, 30+ tables

Predicting protein-ligand binding affinity is an essential part of computer-aided drug design. However, generalisable and performant global binding affinity models remain elusive, particularly in low data regimes. Despite the evolution of model architectures, current benchmarks are not well-suited to probe the generalisability of 3D binding affinity models. Furthermore, 3D global architectures such as GNNs have not lived up to performance expectations. To investigate these issues, we introduce a novel split of the PDBBind dataset, minimizing similarity leakage between train and test sets and allowing for a fair and direct comparison between various model architectures. On this low similarity split, we demonstrate that, in general, 3D global models are superior to protein-specific local models in low data regimes. We also demonstrate that the performance of GNNs benefits from three novel contributions: supervised pre-training via quantum mechanical data, unsupervised pre-training via small molecule diffusion, and explicitly modeling hydrogen atoms in the input graph. We believe that this work introduces promising new approaches to unlock the potential of GNN architectures for binding affinity modelling., Comment: 17 pages, 10 figues

Inverse reinforcement learning methods aim to retrieve the reward function of a Markov decision process based on a dataset of expert demonstrations. The commonplace scarcity and heterogeneous sources of such demonstrations can lead to the absorption of spurious correlations in the data by the learned reward function. Consequently, this adaptation often exhibits behavioural overfitting to the expert data set when a policy is trained on the obtained reward function under distribution shift of the environment dynamics. In this work, we explore a novel regularization approach for inverse reinforcement learning methods based on the causal invariance principle with the goal of improved reward function generalization. By applying this regularization to both exact and approximate formulations of the learning task, we demonstrate superior policy performance when trained using the recovered reward functions in a transfer setting

Collective light-matter interactions have been used to control chemistry and energy transfer, yet accessible approaches that combine ab initio methodology with large many-body quantum optical systems are missing due to the fast increase in computational cost for explicit simulations. We introduce an accessible ab initio quantum embedding concept for many-body quantum optical systems that allows to treat the collective coupling of molecular many-body systems effectively in the spirit of macroscopic QED while keeping the rigor of ab initio quantum chemistry for the molecular structure. Our approach fully includes the quantum fluctuations of the polaritonic field and yet remains much simpler and more intuitive than complex embedding approaches such as dynamical mean-field theory. We illustrate the underlying assumptions by comparison to the Tavis--Cummings model. The intuitive application of the quantized embedding approach and its transparent limitations offer a practical framework for the field of ab initio polaritonic chemistry to describe collective effects in realistic molecular ensembles.

The fractional AC Josephson effect is a discerning property of topological superconductivity in hybrid Josephson junctions. Recent experimental observations of missing odd Shapiro steps and half Josephson frequency emission in various materials have sparked significant debate regarding their potential origin in the effect. In this study, we present microwave emission measurements on a resistively shunted Josephson junction based on a HgTe quantum well. We demonstrate that, with significant spurious inductance in the shunt wiring, the experiment operates in a nonlinear dynamic regime characterized by period-doubling. This leads to additional microwave emission peaks at half of the Josephson frequency, $f_J/2$, which can mimic the $4\pi$-periodicity of topological Andreev states. The observed current-voltage characteristics and emission spectra are well-described by a simple RCLSJ model. Furthermore, we show that the nonlinear dynamics of the junction can be controlled using gate voltage, magnetic field, and temperature, with our model accurately reproducing these effects without incorporating any topological attributes. Our observations urge caution in interpreting emission at $f_J/2$ as evidence for gapless Andreev bound states in topological junctions and suggest the appropriate parameter range for future experiments., Comment: 13 pages, 8 figures

The software project kdotpy provides a Python application for simulating electronic band structures of semiconductor devices with $\mathbf{k}\cdot\mathbf{p}$ theory on a lattice. The application implements the widely used Kane model, capable of reliable predictions of transport and optical properties for a large variety of topological and non-topological materials with a zincblende crystal structure. The application automates the tedious steps of simulating band structures. The user inputs the relevant physical parameters on the command line, for example materials and dimensions of the device, magnetic field, and temperature. The program constructs the appropriate matrix Hamiltonian on a discretized lattice of spatial coordinates and diagonalizes it. The physical observables are extracted from the eigenvalues and eigenvectors and saved as output. The program is highly customizable with a large set of configuration options and material parameters., Comment: 143 pages, submission to SciPost Physics Codebases

We consider quasi-interpolation with a main application in radial basis function approximations and compression in this article. Constructing and using these quasi-interpolants, we consider wavelet and compression-type approximations from their linear spaces and provide convergence estimates. The results include an error estimate for nonlinear approximation by quasi-interpolation, results about compression in the space of continuous functions and a pointwise convergence estimate for approximands of low smoothness.

In unsupervised representation learning, models aim to distill essential features from high-dimensional data into lower-dimensional learned representations, guided by inductive biases. Understanding the characteristics that make a good representation remains a topic of ongoing research. Disentanglement of independent generative processes has long been credited with producing high-quality representations. However, focusing solely on representations that adhere to the stringent requirements of most disentanglement metrics, may result in overlooking many high-quality representations, well suited for various downstream tasks. These metrics often demand that generative factors be encoded in distinct, single dimensions aligned with the canonical basis of the representation space. Motivated by these observations, we propose two novel metrics: Importance-Weighted Orthogonality (IWO) and Importance-Weighted Rank (IWR). These metrics evaluate the mutual orthogonality and rank of generative factor subspaces. Throughout extensive experiments on common downstream tasks, over several benchmark datasets and models, IWO and IWR consistently show stronger correlations with downstream task performance than traditional disentanglement metrics. Our findings suggest that representation quality is closer related to the orthogonality of independent generative processes rather than their disentanglement, offering a new direction for evaluating and improving unsupervised learning models.

To achieve state-of-the-art jet energy resolution for Particle Flow, sophisticated energy clustering algorithms must be developed that can fully exploit available information to separate energy deposits from charged and neutral particles. Three published neural network-based shower separation models were applied to simulation and experimental data to measure the performance of the highly granular CALICE Analogue Hadronic Calorimeter (AHCAL) technological prototype in distinguishing the energy deposited by a single charged and single neutral hadron for Particle Flow. The performance of models trained using only standard spatial and energy and charged track position information from an event was compared to models trained using timing information available from AHCAL, which is expected to improve sensitivity to shower development and, therefore, aid in clustering. Both simulation and experimental data were used to train and test the models and their performances were compared. The best-performing neural network achieved significantly superior event reconstruction when timing information was utilised in training for the case where the charged hadron had more energy than the neutral one, motivating temporally sensitive calorimeters. All models under test were observed to tend to allocate energy deposited by the more energetic of the two showers to the less energetic one. Similar shower reconstruction performance was observed for a model trained on simulation and applied to data and a model trained and applied to data.

A new generalization of shifted thin plate splines $$\varphi(x)=(c^{2d}+||x||^{2d})\log\left(c^{2d}+||x||^{2d}\right),\qquad x\in\mathbb{R}^n, d\in \mathbb{N}, c>0$$ is presented to increase the accuracy of quasi-interpolation further. With the restriction to Euclidean spaces of even dimensionality, the generalization can be used to generate a quasi-Lagrange operator that reproduces all polynomials of degree $n+2d-1$. It thus complements the case of the newly proposed generalized multiquadric $\varphi(x)=\sqrt{c^{2d}+||x||^{2d}},\quad x\in\mathbb{R}^n, d\in \mathbb{N}, c>0$, which is restricted to odd dimensions \cite{ortmann}. This generalization improves the approximation order by a factor of $\mathcal{O}\left(h^{2(d-1)}\right)$, where $d=1$ represents the classical thin plate spline. The results are then compared with the theoretical optimal approximation from the shift-invariant space generated by these functions. Moreover, we introduce a new class of inverse multiquadrics $$\varphi(x)=\left(c^\lambda +||x||^\lambda\right)^\beta,\qquad x\in\mathbb{R}^n, \lambda \in\mathbb{R},\beta \in \mathbb{R}\backslash\mathbb{N}, c>0. $$ We provide an explicit representation of the generalized Fourier transform and discuss its asymptotic behaviour near the origin. Particular emphasis is placed on the case where $\lambda$ and $\beta$ are both negative. It is demonstrated that, in dimensions $n\geq3$, it is possible to build a quasi-Lagrange operator that reproduces all polynomials of degree $n-3$ when $n$ is even and of degree $\frac{n-1}{2}$ when n is odd. Furthermore, the uniform approximation error is given by $\mathcal{O}\left(h^{n-2}\log(1/h)\right)$ for $n$ even and $\mathcal{O}\left(h^{\frac{n-3}{2}}\right)$ for $n$ odd. Here, $h>0$ denotes the fill distance.

The Purcell effect describes the modification of the spontaneous decay rate in the presence of electromagnetic media and bodies. In this work, we shed light on the dependencies and magnitude of this effect for chiral materials. Using the framework of macroscopic quantum electrodynamics and Fermi's golden rule, we study a chiral bulk medium with and without local-field corrections, an idealised chiral mirror and a chiral surface. The results imply that the chiral effect is greatest for large transition frequencies, molecules with large optical rotatory strength and media with a strong cross-susceptibility. In the case of a half space, short distances from the molecule to the interface additionally enhance the effect., Comment: 14 pages, 2 figures, Submitted

Crop breeding is crucial in improving agricultural productivity while potentially decreasing land usage, greenhouse gas emissions, and water consumption. However, breeding programs are challenging due to long turnover times, high-dimensional decision spaces, long-term objectives, and the need to adapt to rapid climate change. This paper introduces the use of Reinforcement Learning (RL) to optimize simulated crop breeding programs. RL agents are trained to make optimal crop selection and cross-breeding decisions based on genetic information. To benchmark RL-based breeding algorithms, we introduce a suite of Gym environments. The study demonstrates the superiority of RL techniques over standard practices in terms of genetic gain when simulated in silico using real-world genomic maize data., Comment: NeurIPS 2023 Workshop on Tackling Climate Change with Machine Learning

Deformable Image Registration (DIR) is essential for aligning medical images that exhibit anatomical variations, facilitating applications such as disease tracking and radiotherapy planning. While classical iterative methods and deep learning approaches have achieved success in DIR, they are often hindered by computational inefficiency or poor generalization. In this paper, we introduce GaussianDIR, a novel, case-specific optimization DIR method inspired by 3D Gaussian splatting. In general, GaussianDIR represents image deformations using a sparse set of mobile and flexible Gaussian primitives, each defined by a center position, covariance, and local rigid transformation. This compact and explicit representation reduces noise and computational overhead while improving interpretability. Furthermore, the movement of individual voxel is derived via blending the local rigid transformation of the neighboring Gaussian primitives. By this, GaussianDIR captures both global smoothness and local rigidity as well as reduces the computational burden. To address varying levels of deformation complexity, GaussianDIR also integrates an adaptive density control mechanism that dynamically adjusts the density of Gaussian primitives. Additionally, we employ multi-scale Gaussian primitives to capture both coarse and fine deformations, reducing optimization to local minima. Experimental results on brain MRI, lung CT, and cardiac MRI datasets demonstrate that GaussianDIR outperforms existing DIR methods in both accuracy and efficiency, highlighting its potential for clinical applications. Finally, as a training-free approach, it challenges the stereotype that iterative methods are inherently slow and transcend the limitations of poor generalization.

The limited robustness of 3D Gaussian Splatting (3DGS) to motion blur and camera noise, along with its poor real-time performance, restricts its application in robotic SLAM tasks. Upon analysis, the primary causes of these issues are the density of views with motion blur and the cumulative errors in dense pose estimation from calculating losses based on noisy original images and rendering results, which increase the difficulty of 3DGS rendering convergence. Thus, a cutting-edge 3DGS-based SLAM system is introduced, leveraging the efficiency and flexibility of 3DGS to achieve real-time performance while remaining robust against sensor noise, motion blur, and the challenges posed by long-session SLAM. Central to this approach is the Fusion Bridge module, which seamlessly integrates tracking-centered ORB Visual Odometry with mapping-centered online 3DGS. Precise pose initialization is enabled by this module through joint optimization of re-projection and rendering loss, as well as strategic view selection, enhancing rendering convergence in large-scale scenes. Extensive experiments demonstrate state-of-the-art rendering quality and localization accuracy, positioning this system as a promising solution for real-world robotics applications that require stable, near-real-time performance. Our project is available at https://ZeldaFromHeaven.github.io/TAMBRIDGE/

We investigate the properties of the photon Bose-Einstein condensate in the limit of small mode spacing. Alongside the well-known threshold of the phase transition at large mode spacings, we find an emergence of a second threshold for sufficiently small mode spacings, defining the crossover to a fully condensed state. Furthermore, we present our findings for the mode occupations in the precondensate or supercooling region towards the continuum limit., Comment: 7 pages, 6 figures

Our study delves into the modifications observed in Vavilov-\v{C}erenkov radiation when its generating charged particle moves parallel to an interface formed by two generic magnetoelectric media, focusing on topological insulators. We compute the electromagnetic field through the Green's function. Applying the far-field approximation and the steepest descent method, we derive analytical expressions for the electric field, revealing contributions from spherical and lateral waves with topological origins. Subsequently, we analyze the angular distribution of the radiation, particularly focusing on parallel motions in close proximity to the interface. Our findings indicate that the radiation along the Vavilov-\v{C}erenkov cone is inhomogeneous and asymmetric. We analyze the radiated energy at both sides of the interface. Finally, we discuss the particle's retarding force, which is notably enhanced in the ultrarelativistic regime. We illustrate these results for the topological insulator TlBiSe$_2$ and the magnetoelectric TbPO$_4$., Comment: 22 pages, 6 figures, Submitted

Motion information from 4D medical imaging offers critical insights into dynamic changes in patient anatomy for clinical assessments and radiotherapy planning and, thereby, enhances the capabilities of 3D image analysis. However, inherent physical and technical constraints of imaging hardware often necessitate a compromise between temporal resolution and image quality. Frame interpolation emerges as a pivotal solution to this challenge. Previous methods often suffer from discretion when they estimate the intermediate motion and execute the forward warping. In this study, we draw inspiration from fluid mechanics to propose a novel approach for continuously modeling patient anatomic motion using implicit neural representation. It ensures both spatial and temporal continuity, effectively bridging Eulerian and Lagrangian specifications together to naturally facilitate continuous frame interpolation. Our experiments across multiple datasets underscore the method's superior accuracy and speed. Furthermore, as a case-specific optimization (training-free) approach, it circumvents the need for extensive datasets and addresses model generalization issues.

Advances in latent diffusion models (LDMs) have revolutionized high-resolution image generation, but the design space of the autoencoder that is central to these systems remains underexplored. In this paper, we introduce LiteVAE, a family of autoencoders for LDMs that leverage the 2D discrete wavelet transform to enhance scalability and computational efficiency over standard variational autoencoders (VAEs) with no sacrifice in output quality. We also investigate the training methodologies and the decoder architecture of LiteVAE and propose several enhancements that improve the training dynamics and reconstruction quality. Our base LiteVAE model matches the quality of the established VAEs in current LDMs with a six-fold reduction in encoder parameters, leading to faster training and lower GPU memory requirements, while our larger model outperforms VAEs of comparable complexity across all evaluated metrics (rFID, LPIPS, PSNR, and SSIM).

Background and purpose: Deformable image registration (DIR) is a crucial tool in radiotherapy for extracting and modelling organ motion. However, when significant changes and sliding boundaries are present, it faces compromised accuracy and uncertainty, determining the subsequential contour propagation and dose accumulation procedures. Materials and methods: We propose an implicit neural representation (INR)-based approach modelling motion continuously in both space and time, named Continues-sPatial-Temporal DIR (CPT-DIR). This method uses a multilayer perception (MLP) network to map 3D coordinate (x,y,z) to its corresponding velocity vector (vx,vy,vz). The displacement vectors (dx,dy,dz) are then calculated by integrating velocity vectors over time. The MLP's parameters can rapidly adapt to new cases without pre-training, enhancing optimisation. The DIR's performance was tested on the DIR-Lab dataset of 10 lung 4DCT cases, using metrics of landmark accuracy (TRE), contour conformity (Dice) and image similarity (MAE). Results: The proposed CPT-DIR can reduce landmark TRE from 2.79mm to 0.99mm, outperforming B-splines' results for all cases. The MAE of the whole-body region improves from 35.46HU to 28.99HU. Furthermore, CPT-DIR surpasses B-splines for accuracy in the sliding boundary region, lowering MAE and increasing Dice coefficients for the ribcage from 65.65HU and 90.41% to 42.04HU and 90.56%, versus 75.40HU and 89.30% without registration. Meanwhile, CPT-DIR offers significant speed advantages, completing in under 15 seconds compared to a few minutes with the conventional B-splines method. Conclusion: Leveraging the continuous representations, the CPT-DIR method significantly enhances registration accuracy, automation and speed, outperforming traditional B-splines in landmark and contour precision, particularly in the challenging areas.

With the emergence of large-scale models trained on diverse datasets, in-context learning has emerged as a promising paradigm for multitasking, notably in natural language processing and image processing. However, its application in 3D point cloud tasks remains largely unexplored. In this work, we introduce Point-In-Context (PIC), a novel framework for 3D point cloud understanding via in-context learning. We address the technical challenge of effectively extending masked point modeling to 3D point clouds by introducing a Joint Sampling module and proposing a vanilla version of PIC called Point-In-Context-Generalist (PIC-G). PIC-G is designed as a generalist model for various 3D point cloud tasks, with inputs and outputs modeled as coordinates. In this paradigm, the challenging segmentation task is achieved by assigning label points with XYZ coordinates for each category; the final prediction is then chosen based on the label point closest to the predictions. To break the limitation by the fixed label-coordinate assignment, which has poor generalization upon novel classes, we propose two novel training strategies, In-Context Labeling and In-Context Enhancing, forming an extended version of PIC named Point-In-Context-Segmenter (PIC-S), targeting improving dynamic context labeling and model training. By utilizing dynamic in-context labels and extra in-context pairs, PIC-S achieves enhanced performance and generalization capability in and across part segmentation datasets. PIC is a general framework so that other tasks or datasets can be seamlessly introduced into our PIC through a unified data format. We conduct extensive experiments to validate the versatility and adaptability of our proposed methods in handling a wide range of tasks and segmenting multi-datasets. Our PIC-S is capable of generalizing unseen datasets and performing novel part segmentation by customizing prompts., Comment: Project page: https://fanglaosi.github.io/Point-In-Context_Pages. arXiv admin note: text overlap with arXiv:2306.08659

We present overlapping top gate electrodes for the formation of gate defined lateral junctions in semiconducting layers as an alternative to the back gate/top gate combination and to the split gate configuration. The optical lithography microfabrication of the overlapping top gates is based on multiple layers of low-temperature atomic layer deposited hafnium oxide, which acts as a gate dielectric and as a robust insulating layer between two overlapping gate electrodes exhibiting a large dielectric breakdown field of > 1E9 V/m. The advantage of overlapping gates over the split gate approach is confirmed in model calculations of the electrostatics of the gate stack. The overlapping gate process is applied to Hall bar devices of mercury telluride in order to study the interaction of different quantum Hall states in the nn', np, pn and pp' regime., Comment: 7 pages, 3 figures

The landscape of traditional industrial manufacturing is undergoing a pivotal shift from resource-intensive production and long supply chains to more sustainable and regionally focused economies. In this evolving scenario, the move towards local, on-demand manufacturing is emerging as a remedy to the environmentally damaging practice of mass-producing products in distant countries and then transporting them over long distances to customers. This paradigm shift significantly empowers customers, giving them greater control over the manufacturing process by enabling on-demand production and favouring local production sites over traditional mass production and extensive shipping practices. In this position paper we propose a cloud-native Manufacturing as a Service (MaaS) platform that integrates advances in three-dimensional (3D) printing technology into a responsive and eco-conscious manufacturing ecosystem. In this context, we propose a high-level architectural design for a cloud-based MaaS platform that connects web shops of local stores with small and medium-sized enterprises (SMEs) operating 3D printers. Furthermore, we outline an experimental design, including a cost-benefit analysis, to empirically evaluate the operational effectiveness and economic feasibility in a cloud-based additive manufacturing ecosystem. The proposed cloud-based MaaS platform enables on-demand additive manufacturing and opens up a profit sharing opportunity between different stakeholders., Comment: In Proceedings of the 14th International Conference on Cloud Computing and Services Science (CLOSER 2024). Angers, France

A neural network for software compensation was developed for the highly granular CALICE Analogue Hadronic Calorimeter (AHCAL). The neural network uses spatial and temporal event information from the AHCAL and energy information, which is expected to improve sensitivity to shower development and the neutron fraction of the hadron shower. The neural network method produced a depth-dependent energy weighting and a time-dependent threshold for enhancing energy deposits consistent with the timescale of evaporation neutrons. Additionally, it was observed to learn an energy-weighting indicative of longitudinal leakage correction. In addition, the method produced a linear detector response and outperformed a published control method regarding resolution for every particle energy studied.

Federated Learning (FL) faces significant challenges with domain shifts in heterogeneous data, degrading performance. Traditional domain generalization aims to learn domain-invariant features, but the federated nature of model averaging often limits this due to its linear aggregation of local learning. To address this, we propose a robust framework, coined as hypernetwork-based Federated Fusion (hFedF), using hypernetworks for non-linear aggregation, facilitating generalization to unseen domains. Our method employs client-specific embeddings and gradient alignment techniques to manage domain generalization effectively. Evaluated in both zero-shot and few-shot settings, hFedF demonstrates superior performance in handling domain shifts. Comprehensive comparisons on PACS, Office-Home, and VLCS datasets show that hFedF consistently achieves the highest in-domain and out-of-domain accuracy with reliable predictions. Our study contributes significantly to the under-explored field of Federated Domain Generalization (FDG), setting a new benchmark for performance in this area.

The band inversion of topological materials in three spatial dimensions is intimately connected to the parity anomaly of two-dimensional massless Dirac fermions. At finite magnetic fields, the parity anomaly reveals itself as a non-zero spectral asymmetry, i.e., a non-zero difference between the number of conduction and valence band Landau levels, due to the unpaired zero Landau level. Here, we realize this two-dimensional Dirac physics at a single surface of the three-dimensional topological insulator (Hg,Mn)Te. We observe an unconventional re-entrant quantum Hall effect that can be directly related to the occurrence of spectral asymmetry in a single topological surface state. The effect should be observable in any topological insulator where the transport is dominated by a single Dirac surface state., Comment: 13 pages, 10 figures, originally submitted version, includes supplementary material

Intra-fraction motion in radiotherapy is commonly modeled using deformable image registration (DIR). However, existing methods often struggle to balance speed and accuracy, limiting their applicability in clinical scenarios. This study introduces a novel approach that harnesses Neural Graphics Primitives (NGP) to optimize the displacement vector field (DVF). Our method leverages learned primitives, processed as splats, and interpolates within space using a shallow neural network. Uniquely, it enables self-supervised optimization at an ultra-fast speed, negating the need for pre-training on extensive datasets and allowing seamless adaptation to new cases. We validated this approach on the 4D-CT lung dataset DIR-lab, achieving a target registration error (TRE) of 1.15\pm1.15 mm within a remarkable time of 1.77 seconds. Notably, our method also addresses the sliding boundary problem, a common challenge in conventional DIR methods.

Background: In MR-guided proton therapy planning, aligning MR and CT images is key for MR-based CT synthesis, especially in mobile regions like the head-and-neck. Misalignments here can lead to less accurate synthetic CT (sCT) images, impacting treatment precision. Purpose: This study introduces a novel network that cohesively unifies image generation and registration processes to enhance the quality and anatomical fidelity of sCTs derived from better-aligned MR images. Methods: The approach synergizes a generation network (G) with a deformable registration network (R), optimizing them jointly in MR-to-CT synthesis. This goal is achieved by alternately minimizing the discrepancies between the generated/registered CT images and their corresponding reference CT counterparts. The generation network employs a UNet architecture, while the registration network leverages an implicit neural representation of the Deformable Vector Fields (DVFs). We validated this method on a dataset comprising 60 Head-and-Neck patients, reserving 12 cases for holdout testing. Results: Compared to the baseline Pix2Pix method with MAE 124.95\pm 30.74 HU, the proposed technique demonstrated 80.98\pm 7.55 HU. The unified translation-registration network produced sharper and more anatomically congruent outputs, showing superior efficacy in converting MR images to sCTs. Additionally, from a dosimetric perspective, the plan recalculated on the resulting sCTs resulted in a remarkably reduced discrepancy to the reference proton plans. Conclusions: This study conclusively demonstrates that a holistic MR-based CT synthesis approach, integrating both image-to-image translation and deformable registration, significantly improves the precision and quality of sCT generation, particularly for the challenging body area with varied anatomic changes between corresponding MR and CT.

Anomaly detection (AD) is the machine learning task of identifying highly discrepant abnormal samples by solely relying on the consistency of the normal training samples. Under the constraints of a distribution shift, the assumption that training samples and test samples are drawn from the same distribution breaks down. In this work, by leveraging tools from causal inference we attempt to increase the resilience of anomaly detection models to different kinds of distribution shifts. We begin by elucidating a simple yet necessary statistical property that ensures invariant representations, which is critical for robust AD under both domain and covariate shifts. From this property, we derive a regularization term which, when minimized, leads to partial distribution invariance across environments. Through extensive experimental evaluation on both synthetic and real-world tasks, covering a range of six different AD methods, we demonstrated significant improvements in out-of-distribution performance. Under both covariate and domain shift, models regularized with our proposed term showed marked increased robustness. Code is available at: https://github.com/JoaoCarv/invariant-anomaly-detection.

We introduce the first generative model trained on the JetClass dataset. Our model generates jets at the constituent level, and it is a permutation-equivariant continuous normalizing flow (CNF) trained with the flow matching technique. It is conditioned on the jet type, so that a single model can be used to generate the ten different jet types of JetClass. For the first time, we also introduce a generative model that goes beyond the kinematic features of jet constituents. The JetClass dataset includes more features, such as particle-ID and track impact parameter, and we demonstrate that our CNF can accurately model all of these additional features as well. Our generative model for JetClass expands on the versatility of existing jet generation techniques, enhancing their potential utility in high-energy physics research, and offering a more comprehensive understanding of the generated jets.

In many states of the quantum electromagnetic field, including the vacuum state, entanglement exists between different space-time regions -- even space-like separated ones. These correlations can be harvested and, thereby, detected by quantum systems which locally interact with the field. Here, we propose an experimental implementation of such an entanglement-harvesting scheme which is based on electro-optic sampling (EOS). We demonstrate that state-of-the-art EOS experiments enable one to harvest entanglement from the vacuum field and to study quantum correlations within general THz fields. We further show how Bell nonlocality present in the vacuum field can be probed. Finally, we introduce a novel approach to mitigate shot noise in single-beam EOS configurations. These findings pave the way for experimental inquiries into foundational properties of relativistic quantum field theory, and empower EOS as a diagnostic tool in THz quantum optics.

While conditional diffusion models are known to have good coverage of the data distribution, they still face limitations in output diversity, particularly when sampled with a high classifier-free guidance scale for optimal image quality or when trained on small datasets. We attribute this problem to the role of the conditioning signal in inference and offer an improved sampling strategy for diffusion models that can increase generation diversity, especially at high guidance scales, with minimal loss of sample quality. Our sampling strategy anneals the conditioning signal by adding scheduled, monotonically decreasing Gaussian noise to the conditioning vector during inference to balance diversity and condition alignment. Our Condition-Annealed Diffusion Sampler (CADS) can be used with any pretrained model and sampling algorithm, and we show that it boosts the diversity of diffusion models in various conditional generation tasks. Further, using an existing pretrained diffusion model, CADS achieves a new state-of-the-art FID of 1.70 and 2.31 for class-conditional ImageNet generation at 256$\times$256 and 512$\times$512 respectively., Comment: Published as a conference paper at ICLR 2024

Physics beyond the Standard Model that is resonant in one or more dimensions has been a longstanding focus of countless searches at colliders and beyond. Recently, many new strategies for resonant anomaly detection have been developed, where sideband information can be used in conjunction with modern machine learning, in order to generate synthetic datasets representing the Standard Model background. Until now, this approach was only able to accommodate a relatively small number of dimensions, limiting the breadth of the search sensitivity. Using recent innovations in point cloud generative models, we show that this strategy can also be applied to the full phase space, using all relevant particles for the anomaly detection. As a proof of principle, we show that the signal from the R\&D dataset from the LHC Olympics is findable with this method, opening up the door to future studies that explore the interplay between depth and breadth in the representation of the data for anomaly detection., Comment: 10 pages, 7 figures

Jets at the LHC, typically consisting of a large number of highly correlated particles, are a fascinating laboratory for deep generative modeling. In this paper, we present two novel methods that generate LHC jets as point clouds efficiently and accurately. We introduce \epcjedi, which combines score-matching diffusion models with the Equivariant Point Cloud (EPiC) architecture based on the deep sets framework. This model offers a much faster alternative to previous transformer-based diffusion models without reducing the quality of the generated jets. In addition, we introduce \epcfm, the first permutation equivariant continuous normalizing flow (CNF) for particle cloud generation. This model is trained with {\it flow-matching}, a scalable and easy-to-train objective based on optimal transport that directly regresses the vector fields connecting the Gaussian noise prior to the data distribution. Our experiments demonstrate that \epcjedi and \epcfm both achieve state-of-the-art performance on the top-quark JetNet datasets whilst maintaining fast generation speed. Most notably, we find that the \epcfm model consistently outperforms all the other generative models considered here across every metric. Finally, we also introduce two new particle cloud performance metrics: the first based on the Kullback-Leibler divergence between feature distributions, the second is the negative log-posterior of a multi-model ParticleNet classifier., Comment: 21 pages, 8 figures

Fast simulation of the energy depositions in high-granular detectors is needed for future collider experiments with ever-increasing luminosities. Generative machine learning (ML) models have been shown to speed up and augment the traditional simulation chain in physics analysis. However, the majority of previous efforts were limited to models relying on fixed, regular detector readout geometries. A major advancement is the recently introduced CaloClouds model, a geometry-independent diffusion model, which generates calorimeter showers as point clouds for the electromagnetic calorimeter of the envisioned International Large Detector (ILD). In this work, we introduce CaloClouds II which features a number of key improvements. This includes continuous time score-based modelling, which allows for a 25-step sampling with comparable fidelity to CaloClouds while yielding a $6\times$ speed-up over Geant4 on a single CPU ($5\times$ over CaloClouds). We further distill the diffusion model into a consistency model allowing for accurate sampling in a single step and resulting in a $46\times$ ($37\times$ over CaloClouds) speed-up. This constitutes the first application of consistency distillation for the generation of calorimeter showers., Comment: 30 pages, 7 figures, 3 tables, code available at https://github.com/FLC-QU-hep/CaloClouds-2

A new generalization of multiquadric functions $\phi(x)=\sqrt{c^{2d}+||x||^{2d}}$, where $x\in\mathbb{R}^n$, $c\in \mathbb{R}$, $d\in \mathbb{N}$, is presented to increase the accuracy of quasi-interpolation further. With the restriction to Euclidean spaces of odd dimensionality, the generalization can be used to generate a quasi-Lagrange operator that reproduces all polynomials of degree $2d-1$. In contrast to the classical multiquadric, the convergence rate of the quasi-interpolation operator can be significantly improved by a factor $h^{2d-n-1}$, where $h>0$ represents the grid spacing. Among other things, we compute the generalized Fourier transform of this new multiquadric function. Finally, an infinite regular grid is employed to analyse the properties of the aforementioned generalization in detail.

In this paper, we study functional approximations where we choose the so-called radial basis function method and more specifically, quasi-interpolation. From the various available approaches to the latter, we form new quasi-Lagrange functions when the orders of the singularities of the radial function's Fourier transforms at zero do not match the parity of the dimension of the space, and therefore new expansions and coefficients are needed to overcome this problem. We develop explicit constructions of infinite Fourier expansions that provide these coefficients and make an extensive comparison of the approximation qualities and - with a particular focus - polynomial precision and uniform approximation order of the various formulae. One of the interesting observations concerns the link between algebraic conditions of expansion coefficients and analytic properties of localness and convergence.

Imitation learning methods are used to infer a policy in a Markov decision process from a dataset of expert demonstrations by minimizing a divergence measure between the empirical state occupancy measures of the expert and the policy. The guiding signal to the policy is provided by the discriminator used as part of an versarial optimization procedure. We observe that this model is prone to absorbing spurious correlations present in the expert data. To alleviate this issue, we propose to use causal invariance as a regularization principle for adversarial training of these models. The regularization objective is applicable in a straightforward manner to existing adversarial imitation frameworks. We demonstrate the efficacy of the regularized formulation in an illustrative two-dimensional setting as well as a number of high-dimensional robot locomotion benchmark tasks., Comment: Published at the ICML 2023 Workshop on Spurious Correlations, Invariance, and Stability

Nonlinear planar magnetotransport is ubiquitous in topological HgTe structures, both in tensile (topological insulator) or compressively strained layers (Weyl semimetal phase). We show that the common reason for the nonlinear planar magnetotransport is the presence of tilted Dirac cones combined with the formation of charge puddles. The origin of the tilted Dirac cones is the mix of the Zeeman term due to the in-plane magnetic field and quadratic contributions to the dispersion relation. We develop a network model that mimics transport of tilted Dirac fermions in the landscape of charge puddles. The model captures the essential features of the experimental data. It should be relevant for nonlinear planar magnetotransport in a variety of topological and small band gap materials., Comment: 12 pages, 11 figures, 1 table

Explainable AI aims to render model behavior understandable by humans, which can be seen as an intermediate step in extracting causal relations from correlative patterns. Due to the high risk of possible fatal decisions in image-based clinical diagnostics, it is necessary to integrate explainable AI into these safety-critical systems. Current explanatory methods typically assign attribution scores to pixel regions in the input image, indicating their importance for a model's decision. However, they fall short when explaining why a visual feature is used. We propose a framework that utilizes interpretable disentangled representations for downstream-task prediction. Through visualizing the disentangled representations, we enable experts to investigate possible causation effects by leveraging their domain knowledge. Additionally, we deploy a multi-path attribution mapping for enriching and validating explanations. We demonstrate the effectiveness of our approach on a synthetic benchmark suite and two medical datasets. We show that the framework not only acts as a catalyst for causal relation extraction but also enhances model robustness by enabling shortcut detection without the need for testing under distribution shifts.

With the rise of large-scale models trained on broad data, in-context learning has become a new learning paradigm that has demonstrated significant potential in natural language processing and computer vision tasks. Meanwhile, in-context learning is still largely unexplored in the 3D point cloud domain. Although masked modeling has been successfully applied for in-context learning in 2D vision, directly extending it to 3D point clouds remains a formidable challenge. In the case of point clouds, the tokens themselves are the point cloud positions (coordinates) that are masked during inference. Moreover, position embedding in previous works may inadvertently introduce information leakage. To address these challenges, we introduce a novel framework, named Point-In-Context, designed especially for in-context learning in 3D point clouds, where both inputs and outputs are modeled as coordinates for each task. Additionally, we propose the Joint Sampling module, carefully designed to work in tandem with the general point sampling operator, effectively resolving the aforementioned technical issues. We conduct extensive experiments to validate the versatility and adaptability of our proposed methods in handling a wide range of tasks., Comment: Project page: https://github.com/fanglaosi/Point-In-Context

Source radiation (radiation reaction) and vacuum-field fluctuations can be seen as two inseparable contributions to processes such as spontaneous emission, the Lamb shift, or the Casimir force. Here, we propose how they can be individually probed and their space-time structure revealed in electro-optic sampling experiments. This allows to experimentally study causality at the single photon level and to reveal space- and time-like correlations in the quantum vacuum. A connection to the time-domain fluctuation-dissipation theorem is also made., Comment: 24 pages, 4 figures

Simulating showers of particles in highly-granular detectors is a key frontier in the application of machine learning to particle physics. Achieving high accuracy and speed with generative machine learning models would enable them to augment traditional simulations and alleviate a major computing constraint. This work achieves a major breakthrough in this task by, for the first time, directly generating a point cloud of a few thousand space points with energy depositions in the detector in 3D space without relying on a fixed-grid structure. This is made possible by two key innovations: i) Using recent improvements in generative modeling we apply a diffusion model to generate photon showers as high-cardinality point clouds. ii) These point clouds of up to $6,000$ space points are largely geometry-independent as they are down-sampled from initial even higher-resolution point clouds of up to $40,000$ so-called Geant4 steps. We showcase the performance of this approach using the specific example of simulating photon showers in the planned electromagnetic calorimeter of the International Large Detector (ILD) and achieve overall good modeling of physically relevant distributions., Comment: 25 pages, 11 figures

The upgrade of the CMS experiment for the high luminosity operation of the LHC comprises the replacement of the current endcap calorimeter by a high granularity sampling calorimeter (HGCAL). The electromagnetic section of the HGCAL is based on silicon sensors interspersed between lead and copper (or copper tungsten) absorbers. The hadronic section uses layers of stainless steel as an absorbing medium and silicon sensors as an active medium in the regions of high radiation exposure, and scintillator tiles directly readout by silicon photomultipliers in the remaining regions. As part of the development of the detector and its readout electronic components, a section of a silicon-based HGCAL prototype detector along with a section of the CALICE AHCAL prototype was exposed to muons, electrons and charged pions in beam test experiments at the H2 beamline at the CERN SPS in October 2018. The AHCAL uses the same technology as foreseen for the HGCAL but with much finer longitudinal segmentation. The performance of the calorimeters in terms of energy response and resolution, longitudinal and transverse shower profiles is studied using negatively charged pions, and is compared to GEANT4 predictions. This is the first report summarizing results of hadronic showers measured by the HGCAL prototype using beam test data., Comment: Accepted for publication by JINST

High-frequency transport in the edge states of the quantum spin Hall (QSH) effect has to date rarely been explored, though it could cast light on the scattering mechanisms taking place therein. We here report on the measurement of the plasmon velocity in topological HgTe quantum wells both in the QSH and quantum Hall (QH) regimes, using harmonic GHz excitations and phase-resolved detection. We observe low plasmon velocities corresponding to large transverse widths, which we ascribe to the prominent influence of charge puddles forming in the vicinity of edge channels. Together with other recent works, it suggests that puddles play an essential role in the edge state physics and probably constitute a main hurdle on the way to clean and robust edge transport., Comment: 8 pages, 6 figures, + supplementary material

With the vast data-collecting capabilities of current and future high-energy collider experiments, there is an increasing demand for computationally efficient simulations. Generative machine learning models enable fast event generation, yet so far these approaches are largely constrained to fixed data structures and rigid detector geometries. In this paper, we introduce EPiC-GAN - equivariant point cloud generative adversarial network - which can produce point clouds of variable multiplicity. This flexible framework is based on deep sets and is well suited for simulating sprays of particles called jets. The generator and discriminator utilize multiple EPiC layers with an interpretable global latent vector. Crucially, the EPiC layers do not rely on pairwise information sharing between particles, which leads to a significant speed-up over graph- and transformer-based approaches with more complex relation diagrams. We demonstrate that EPiC-GAN scales well to large particle multiplicities and achieves high generation fidelity on benchmark jet generation tasks., Comment: 18 pages, 8 figures, 3 tables, code available at: https://github.com/uhh-pd-ml/EPiC-GAN

HgTe is a versatile topological material and has enabled the realization of a variety of topological states, including two- and three-dimensional (3D) topological insulators and topological semimetals. Nevertheless, a quantitative understanding of its electronic structure remains challenging, in particular due to coupling of the Te 5p-derived valence electrons to Hg 5d core states at shallow binding energy. We present a joint experimental and theoretical study of the electronic structure in strained HgTe(001) films in the 3D topological-insulator regime, based on angle-resolved photoelectron spectroscopy and density functional theory. The results establish detailed agreement in terms of (i) electronic band dispersions and orbital symmetries, (ii) surface and bulk contributions to the electronic structure, and (iii) the importance of Hg 5d states in the valence-band formation. Supported by theory, our experiments directly image the paradigmatic band inversion in HgTe, underlying its non-trivial band topology.