Search

A force field as accurate as quantum mechanics (QM) and as fast as molecular mechanics (MM), with which one can simulate a biomolecular system efficiently enough and meaningfully enough to get quantitative insights, is among the most ardent dreams of biophysicists -- a dream, nevertheless, not to be fulfilled any time soon. Machine learning force fields (MLFFs) represent a meaningful endeavor towards this direction, where differentiable neural functions are parametrized to fit ab initio energies, and furthermore forces through automatic differentiation. We argue that, as of now, the utility of the MLFF models is no longer bottlenecked by accuracy but primarily by their speed (as well as stability and generalizability), as many recent variants, on limited chemical spaces, have long surpassed the chemical accuracy of $1$ kcal/mol -- the empirical threshold beyond which realistic chemical predictions are possible -- though still magnitudes slower than MM. Hoping to kindle explorations and designs of faster, albeit perhaps slightly less accurate MLFFs, in this review, we focus our attention on the design space (the speed-accuracy tradeoff) between MM and ML force fields. After a brief review of the building blocks of force fields of either kind, we discuss the desired properties and challenges now faced by the force field development community, survey the efforts to make MM force fields more accurate and ML force fields faster, envision what the next generation of MLFF might look like.

In this paper, we assess the movement error of a targeting system given target location data from artificial intelligence (AI) methods in automatic target recognition (ATR) systems. Few studies evaluate the impacts on the accuracy in moving a targeting system to an aimpoint provided in this manner. To address this knowledge gap, we assess the performance of a controlled gun turret system given target location from an object detector developed from AI methods. In our assessment, we define a measure of object detector error and examine the correlations with several standard metrics in object detection. We then statistically analyze the object detector error data and turret movement error data acquired from controlled targeting simulations, as well as their aggregate error, to examine the impact on turret movement accuracy. Finally, we study the correlations between additional metrics and the probability of a hit. The results indicate that AI technologies are a significant source of error to targeting systems. Moreover, the results suggest that metrics such as the confidence score, intersection-over-union, average precision and average recall are predictors of accuracy against stationary targets with our system parameters., Comment: 30 pages, 21 figures

The purpose of this paper is to characterize aiming errors in controlled weapon systems given target location as input. To achieve this objective, we analyze the accuracy of a controlled weapon system model for stationary and moving targets under different error sources and firing times. First, we develop a mathematical model of a gun turret and use it to design two controllers, a Proportional-Integral-Derivative controller and a Model Predictive controller, which accept the target location input and move the turret to the centroid of the target in simulations. For stationary targets, we analyze the impact of errors in estimating the system's parameters and uncertainty in the aim point measurement. Our results indicate that turret movement is more sensitive to errors in the moment of inertia than the damping coefficient, which could lead to incorrect simulations of controlled turret system accuracy. The results also support the hypothesis that turret movement errors are larger over longer distances of gun turret movement and, assuming no time constraints, accuracy improves the longer one waits to fire; though this may not always be practical in a combat scenario. Additionally, we demonstrate that the integral control component is needed for high accuracy in moving target scenarios., Comment: 29 pages, 15 figures

Modern interferometers such as LIGO have achieved sensitivities limited by quantum noise, comprised of radiation pressure and shot noise. To mitigate this noise, a static system is employed that minimizes the quantum noise within the measurement band. However, since gravitational wave inspiral signals are a single frequency changing over time, only noise at the chirp frequency needs to be minimized. Here we demonstrate dynamically tracking a target signal using an optical spring, resulting in an increased signal to noise ratio (SNR). We report on a SNR increase by up to a factor of 40 when compared to a static configuration.

Low-mass galaxies provide a powerful tool with which to investigate departures from the standard cosmological paradigm in models that suppress the abundance of small dark matter structures. One of the simplest metrics that can be used to compare different models is the abundance of satellite galaxies in the Milky Way. Viable dark matter models must produce enough substructure to host the observed number of Galactic satellites. Here, we scrutinize the predictions of the neutrino Minimal Standard Model ($\nu{\rm MSM}$), a well-motivated extension of the Standard Model of particle physics in which the production of sterile neutrino dark matter is resonantly enhanced by a lepton asymmetry in the primordial plasma. This process enables the model to evade current constraints associated with non-resonantly produced dark matter. Independently of assumptions about galaxy formation physics we rule out, with at least 95 per cent confidence, all parameterizations of the $\nu{\rm MSM}$ with sterile neutrino rest mass, $M_{\rm s} \leq 1.4\, {\rm keV}$. Incorporating physically motivated prescriptions of baryonic processes and modelling the effects of reionization strengthen our constraints, and we exclude all $\nu{\rm MSM}$ parameterizations with $M_{\rm s} \leq 4\, {\rm keV}$. Unlike other literature, our fiducial constraints do not rule out the putative 3.55 keV X-ray line, if it is indeed produced by the decay of a sterile neutrino; however, some of the most favoured parameter space is excluded. If the Milky Way satellite count is higher than we assume, or if the Milky Way halo is less massive than $M^{\rm MW}_{200} = 8 \times 10^{11}\, {\rm M_\odot}$, we rule out the $\nu{\rm MSM}$ as the origin of the 3.55 keV excess. In contrast with other work, we find that the constraints from satellite counts are substantially weaker than those reported from X-ray non-detections., Comment: 14 pages, six figures, two appendices. A repository of reduced scripts and data will be made available at https://github.com/Musical-Neutron/nuMSM_constraints. Submitted, comments welcome!

Combining neuroimaging datasets from multiple sites and scanners can help increase statistical power and thus provide greater insight into subtle neuroanatomical effects. However, site-specific effects pose a challenge by potentially obscuring the biological signal and introducing unwanted variance. Existing harmonization techniques, which use statistical models to remove such effects, have been shown to incompletely remove site effects while also failing to preserve biological variability. More recently, generative models using GANs or autoencoder-based approaches, have been proposed for site adjustment. However, such methods are known for instability during training or blurry image generation. In recent years, diffusion models have become increasingly popular for their ability to generate high-quality synthetic images. In this work, we introduce the disentangled diffusion autoencoder (DDAE), a novel diffusion model designed for controlling specific aspects of an image. We apply the DDAE to the task of harmonizing MR images by generating high-quality site-adjusted images that preserve biological variability. We use data from 7 different sites and demonstrate the DDAE's superiority in generating high-resolution, harmonized 2D MR images over previous approaches. As far as we are aware, this work marks the first diffusion-based model for site adjustment of neuroimaging data.

Cooling of atomic motion is a crucial tool for many branches of atomic physics, ranging from fundamental physics explorations to quantum information and sensing. For trapped ions, electromagnetically-induced-transparency (EIT) cooling has received attention for the relative speed, low laser power requirements, and broad cooling bandwidth of the technique. However, in applications where the ion used for cooling has hyperfine structure to enable long coherence times, it is difficult to find a closed three-level system in which to perform standard EIT cooling. Here, we demonstrate successful EIT cooling on 25Mg+ by the addition of an extra laser frequency; this method can be applied to any ion with non-zero nuclear spin. Furthermore, we demonstrate simultaneous EIT cooling of all axial modes in mixed-species crystals 9Be+ - 25Mg+ and 9Be+ - 25Mg+ - 9Be+ through the 25Mg+ ion., Comment: 10 pages, 6 figures, 2 tables

How do complex adaptive systems, such as life, emerge from simple constituent parts? In the 1990s Walter Fontana and Leo Buss proposed a novel modeling approach to this question, based on a formal model of computation known as $\lambda$ calculus. The model demonstrated how simple rules, embedded in a combinatorially large space of possibilities, could yield complex, dynamically stable organizations, reminiscent of biochemical reaction networks. Here, we revisit this classic model, called AlChemy, which has been understudied over the past thirty years. We reproduce the original results and study the robustness of those results using the greater computing resources available today. Our analysis reveals several unanticipated features of the system, demonstrating a surprising mix of dynamical robustness and fragility. Specifically, we find that complex, stable organizations emerge more frequently than previously expected, that these organizations are robust against collapse into trivial fixed-points, but that these stable organizations cannot be easily combined into higher order entities. We also study the role played by the random generators used in the model, characterizing the initial distribution of objects produced by two random expression generators, and their consequences on the results. Finally, we provide a constructive proof that shows how an extension of the model, based on typed $\lambda$ calculus, could simulate transitions between arbitrary states in any possible chemical reaction network, thus indicating a concrete connection between AlChemy and chemical reaction networks. We conclude with a discussion of possible applications of AlChemy to self-organization in modern programming languages and quantitative approaches to the origin of life.

The assessment of mitral regurgitation (MR) using cardiac MRI, particularly Cine MRI, is a promising technique due to its wide availability. However, some of the temporal information available in clinical Cine MRI may not be fully utilised, as it requires detailed temporal analysis across different cardiac views. We propose a new approach to identify MR which automatically extracts 4-dimensional (3D + Time) morphological features from the reconstructed mitral annulus (MA) using Cine long-axis (LAX) views MRI. Our feature extraction involves locating the MA insertion points to derive the reconstructed MA geometry and displacements, resulting in a total of 187 candidate features. We identify the 25 most relevant mitral valve features using minimum-redundancy maximum-relevance (MRMR) feature selection technique. We then apply linear discriminant analysis (LDA) and random forest (RF) model to determine the presence of MR. Both LDA and RF demonstrate good performance, with accuracies of 0.72 +/- 0.05 and 0.73 +/- 0.09, respectively, in a 5-fold cross-validation analysis. This approach will be incorporated in an automatic tool to identify valvular diseases from Cine MRI by integrating both handcrafted and deep features. Our tool will facilitate the diagnosis of valvular disease from conventional cardiac MRI scans with no additional scanning or image analysis penalty. All code is made available on an open-source basis at: https://github.com/HenryOn2021/MA_Morphological_Features., Comment: Accepted paper in STACOM 2024

The jet cross-section and jet-substructure observables in $p$$+$$p$ collisions at $\sqrt{s}=200$ GeV were measured by the PHENIX Collaboration at the Relativistic Heavy Ion Collider (RHIC). Jets are reconstructed from charged-particle tracks and electromagnetic-calorimeter clusters using the anti-$k_{t}$ algorithm with a jet radius $R=0.3$ for jets with transverse momentum within $8.0

In this paper, an application of a 1D deep convolutional neural network (DCNN) and 4x4 1D DCNN Multi-channel Model (DCNN-MCM) was developed to predict the probability of a channel being associated with a given transmitter for each emitter in a 4x4 multi-input multi-output (MIMO) system. Counterintuitively, compared to the traditional approach to emitter association (EA), this research argues for the identification of received RF signals based on channel features (CFs) such as the channel impulse response (CIR) and transfer function (TF). Based on the CFs there are unique properties per transmit and receive pair that can be used to identify the different transmitters. More specifically, CFs are often defined by a wide sense stationary (WSS) stochastic process which gives them unique properties such as, being statistically independent and reciprocal for emitter association. Given these CFs, the received RF signals can be used to measure the clutter and objects surrounding the emitters. Safely assuming the surrounding clutter and objects are unique to each emitter, the estimation and tracking of this clutter provides a way of identifying what emitter is transmitting the observed signal. This method successfully classifies and identifies the emitters at 97.22% and 88.89% accuracy for the DCNN and DCNN-MCM respectively. This method was further compared with the accuracy of previous methods to test the validity of the approach and future improvements.

Quantum computers must achieve large-scale, fault-tolerant operation to deliver on their promise of transformational processing power [1-4]. This will require thousands or millions of high-fidelity quantum gates and similar numbers of qubits [5]. Demonstrations using neutral-atom qubits trapped and manipulated by lasers have shown that this modality can provide high two-qubit gate (CZ) fidelities and scalable operation [6-10]. However, the gates in these demonstrations are driven by lasers that do not resolve individual qubits, with universal computation enabled by physical mid-circuit shuttling of the qubits. This relatively slow operation will greatly extend runtimes for useful, large-scale computation. Here we demonstrate a universal neutral-atom quantum computer with gate rates limited by optical switching times, rather than shuttling, by individually addressing tightly focused laser beams at an array of single atoms. We achieve CZ fidelity of 99.35(4)% and local single qubit RZ gate fidelity of 99.902(8)%. Moreover, we demonstrate non-destructive readout of alkali-atom qubits with sub-percent loss, which boosts operational speed. This technique also enables us to measure 99.73(3)% CZ fidelity with atom-loss events excluded, which is a record among long lived neutral-atom qubits and highlights the path to higher fidelity and error correction. Our results represent a critical step towards large-scale, fault-tolerant neutral-atom quantum computers that can execute computations on practical timescales.

This paper introduces an approach to improve volume conservation in the immersed boundary (IB) method using regularized delta functions derived from composite B-splines. These delta functions employ tensor product kernels using B-splines, whose polynomial degrees vary in normal and tangential directions based on the corresponding velocity component. Our method addresses the long-standing volume conservation issues in the conventional IB method, particularly evident in simulations of pressurized, closed membranes. We demonstrate that our approach significantly enhances volume conservation, rivaling the performance of the non-local Divergence-Free Immersed Boundary (DFIB) method introduced by Bao et al. while maintaining the local nature of the classical IB method. This avoids the computational overhead associated with the DFIB method's construction of an explicit velocity potential which requires additional Poisson solves. Numerical experiments show that sufficiently regular composite B-spline kernels can maintain initial volumes to within machine precision. We analyze the relationship between kernel regularity and the accuracy of force spreading and velocity interpolation operations. Our findings indicate that composite B-splines of at least $C^1$ regularity produce results comparable to the DFIB method in dynamic simulations, with volume conservation errors primarily dominated by the time-stepping scheme's truncation error. This work offers a computationally efficient alternative for improving volume conservation in IB methods, particularly beneficial for large-scale, three-dimensional simulations. The proposed approach requires minimal modifications to an existing IB code, making it an accessible improvement for a wide range of applications in computational fluid dynamics and fluid-structure interaction., Comment: 33 pages, 14 figures

For $0\leq k\leq n-1$, we introduce a family of $k$-skeletal paths which are counted by the $n$-th Catalan number for each $k$, and specialize to Dyck paths when $k=n-1$. We similarly introduce $k$-skeletal parking functions which are equinumerous with the spanning trees on $n+1$ vertices for each $k$, and specialize to classical parking functions for $k=n-1$. The preceding constructions are generalized to paths lying in a trapezoid with base $c > 0$ and southeastern diagonal of slope $1/m$; $c$ and $m$ need not be integers. We give bijections among these families when $k$ varies with $m$ and $c$ fixed. Our constructions are motivated by chip firing and have connections to combinatorial representation theory and tropical geometry., Comment: 29 pages, 9 figures

We introduce Imagen 3, a latent diffusion model that generates high quality images from text prompts. We describe our quality and responsibility evaluations. Imagen 3 is preferred over other state-of-the-art (SOTA) models at the time of evaluation. In addition, we discuss issues around safety and representation, as well as methods we used to minimize the potential harm of our models.

We take the first step towards incorporating compound nuclear observables at astrophysically-relevant energies into the experimental evidence used to constrain optical models, by propagating the uncertainty in two global optical potentials, one phenomenological and one microscopic, to correlated fission observables using the Monte Carlo Hauser-Feshbach formalism. We compare to historic and recent experimental data, and find that the parametric optical model uncertainty in neutron-fragment correlated observables is significant in observables involving neutron energies and multiplicities. Different phenomenological and microscopic potentials disagree, both with experiment and with eachcother, which we attribute to the extrapolation of the phenomenological models away from stability, and to the limitations of the nuclear matter folding approach for the microscopic model. We comment on the role of the optical model in future efforts to globally optimize fission models.

Recent and unremitting capability advances have been accompanied by calls for comprehensive, rather than patchwork, regulation of frontier artificial intelligence (AI). Approval regulation is emerging as a promising candidate. An approval regulation scheme is one in which a firm cannot legally market, or in some cases develop, a product without explicit approval from a regulator on the basis of experiments performed upon the product that demonstrate its safety. This approach is used successfully by the FDA and FAA. Further, its application to frontier AI has been publicly supported by many prominent stakeholders. This report proposes an approval regulation schematic for only the largest AI projects in which scrutiny begins before training and continues through to post-deployment monitoring. The centerpieces of the schematic are two major approval gates, the first requiring approval for large-scale training and the second for deployment. Five main challenges make implementation difficult: noncompliance through unsanctioned deployment, specification of deployment readiness requirements, reliable model experimentation, filtering out safe models before the process, and minimizing regulatory overhead. This report makes a number of crucial recommendations to increase the feasibility of approval regulation, some of which must be followed urgently if such a regime is to succeed in the near future. Further recommendations, produced by this report's analysis, may improve the effectiveness of any regulatory regime for frontier AI.

Soil moisture estimation is an important task to enable precision agriculture in creating optimal plans for irrigation, fertilization, and harvest. It is common to utilize statistical and machine learning models to estimate soil moisture from traditional data sources such as weather forecasts, soil properties, and crop properties. However, there is a growing interest in utilizing aerial and geospatial imagery to estimate soil moisture. Although these images capture high-resolution crop details, they are expensive to curate and challenging to interpret. Imagine, an AI-enhanced software tool that predicts soil moisture using visual cues captured by smartphones and statistical data given by weather forecasts. This work is a first step towards that goal of developing a multi-modal approach for soil moisture estimation. In particular, we curate a dataset consisting of real-world images taken from ground stations and their corresponding weather data. We also propose MIS-ME - Meteorological & Image based Soil Moisture Estimator, a multi-modal framework for soil moisture estimation. Our extensive analysis shows that MIS-ME achieves a MAPE of 10.14%, outperforming traditional unimodal approaches with a reduction of 3.25% in MAPE for meteorological data and 2.15% in MAPE for image data, highlighting the effectiveness of tailored multi-modal approaches. Our code and dataset will be available at https://github.com/OSU-Complex-Systems/MIS-ME.git., Comment: Accepted by DSAA2024

The potential future democratization of spaceflight reveals a need for design of experiences that extend beyond our current conceptualization of spaceflight. Research on career astronauts indicates that transformative experiences occur during spaceflight despite the physiological and psychological stressors involved. This phenomenon allows us to envision a future where such profound experiences are accessible to diverse spaceflight participants. In this position paper, we advocate for acknowledging how design decisions made at the genesis of commercial spaceflight might impact space travelers of this speculative future. In proposing salutogenesis as an orienting topic, a potential design framework, and as a metric for spaceflight participant experience, we offer a call to action for the broader experience design community to engage with the design of profound experiences for spaceflight participants., Comment: SpaceCHI 2023

We resolve the periodic square peg problem using a simple Lagrangian Floer homology argument. Inscribed squares are interpreted as intersections between two non-displaceable Lagrangian sub-manifolds of a symplectic 4-torus.

The interplay of unitary evolution and projective measurements is a modern interest in the study of many-body entanglement. On the one hand, the competition between these two processes leads to the recently-discovered measurement-induced phase transition (MIPT). On the other, measurement-based quantum computation (MBQC) is a well-known computational model studying how measurements simulate unitary evolution utilizing the entanglement of special resources such as the 2D cluster state. The entanglement properties enabling MBQC may be attributed to symmetry-protected topological (SPT) orders, particularly subsystem symmetric (SSPT) orders. It was recently found that the 1D cluster state may be associated with an SPT phase in random circuits respecting a global $Z_2 \times Z_2$ symmetry, and furthermore that all phase transitions in this scenario belong to the same universality class. As resources with greater computational power feature greater symmetry, it is fruitful to investigate further any relationship between levels of symmetry in MIPTs and MBQC. In this paper we investigate MIPTs on a torus with three levels of symmetry-respecting unitary evolution interspersed by measurements. Although we find two area-law phases and one volume-law phase with distinct entanglement structures for each ensemble, the phase transition from the volume-law phase to the area-law phase associated with the 2D cluster state has variable correlation length exponent $\nu$. Whereas $\nu\approx 0.90$ for unconstrained Clifford unitaries and $\nu\approx 0.83$ for globally-symmetric Cliffords, subsystem-symmetric Cliffords feature a much smaller value $\nu\approx 0.38$. It is speculated that the hierarchy of distinct transitions seen in these random monitored quantum circuit models might have consequences for computational universality in MBQC., Comment: 20 pages, 16 figures

In this paper, we reveal that the signal representation of information introduced by Gentzkow and Kamenica (2017) can be applied profitably to dynamic decision problems. We use this to characterize when one dynamic information structure is more valuable to an agent than another, irrespective of what other dynamic sources of information the agent may possess. Notably, this robust dominance is equivalent to an intuitive dynamic version of Brooks, Frankel, and Kamenica (2022)'s reveal-or-refine condition.

General-purpose artificial intelligence (AI) systems are built on massive swathes of public web data, assembled into corpora such as C4, RefinedWeb, and Dolma. To our knowledge, we conduct the first, large-scale, longitudinal audit of the consent protocols for the web domains underlying AI training corpora. Our audit of 14,000 web domains provides an expansive view of crawlable web data and how codified data use preferences are changing over time. We observe a proliferation of AI-specific clauses to limit use, acute differences in restrictions on AI developers, as well as general inconsistencies between websites' expressed intentions in their Terms of Service and their robots.txt. We diagnose these as symptoms of ineffective web protocols, not designed to cope with the widespread re-purposing of the internet for AI. Our longitudinal analyses show that in a single year (2023-2024) there has been a rapid crescendo of data restrictions from web sources, rendering ~5%+ of all tokens in C4, or 28%+ of the most actively maintained, critical sources in C4, fully restricted from use. For Terms of Service crawling restrictions, a full 45% of C4 is now restricted. If respected or enforced, these restrictions are rapidly biasing the diversity, freshness, and scaling laws for general-purpose AI systems. We hope to illustrate the emerging crises in data consent, for both developers and creators. The foreclosure of much of the open web will impact not only commercial AI, but also non-commercial AI and academic research., Comment: 41 pages (13 main), 5 figures, 9 tables

Neuroscience research has made immense progress over the last decade, but our understanding of the brain remains fragmented and piecemeal: the dream of probing an arbitrary brain region and automatically reading out the information encoded in its neural activity remains out of reach. In this work, we build towards a first foundation model for neural spiking data that can solve a diverse set of tasks across multiple brain areas. We introduce a novel self-supervised modeling approach for population activity in which the model alternates between masking out and reconstructing neural activity across different time steps, neurons, and brain regions. To evaluate our approach, we design unsupervised and supervised prediction tasks using the International Brain Laboratory repeated site dataset, which is comprised of Neuropixels recordings targeting the same brain locations across 48 animals and experimental sessions. The prediction tasks include single-neuron and region-level activity prediction, forward prediction, and behavior decoding. We demonstrate that our multi-task-masking (MtM) approach significantly improves the performance of current state-of-the-art population models and enables multi-task learning. We also show that by training on multiple animals, we can improve the generalization ability of the model to unseen animals, paving the way for a foundation model of the brain at single-cell, single-spike resolution.

Predicting survival in Amyotrophic Lateral Sclerosis (ALS) is a challenging task. Magnetic resonance imaging (MRI) data provide in vivo insight into brain health, but the low prevalence of the condition and resultant data scarcity limit training set sizes for prediction models. Survival models are further hindered by the subtle and often highly localised profile of ALS-related neurodegeneration. Normative models present a solution as they increase statistical power by leveraging large healthy cohorts. Separately, diffusion models excel in capturing the semantics embedded within images including subtle signs of accelerated brain ageing, which may help predict survival in ALS. Here, we combine the benefits of generative and normative modelling by introducing the normative diffusion autoencoder framework. To our knowledge, this is the first use of normative modelling within a diffusion autoencoder, as well as the first application of normative modelling to ALS. Our approach outperforms generative and non-generative normative modelling benchmarks in ALS prognostication, demonstrating enhanced predictive accuracy in the context of ALS survival prediction and normative modelling in general.

The surface brightness distribution of massive early-type galaxies (ETGs) often deviates from a perfectly elliptical shape. To capture these deviations in their isophotes during an ellipse fitting analysis, Fourier modes of order $m = 3, 4$ are often used. In such analyses the centre of each ellipse is treated as a free parameter, which may result in offsets from the centre of light, particularly for ellipses in the outer regions. This complexity is not currently accounted for in the mass models used in either strong gravitational lensing or galactic dynamical studies. In this work, we adopt a different approach, using the $m=1$ Fourier mode to account for this complexity while keeping the centres of all perturbed ellipses fixed, showing that it fits the data equally well. We applied our method to the distribution of light emission to a sample of ETGs from the MASSIVE survey and found that the majority have low $m_1$ amplitudes, below 2 percent. Five out of the 30 galaxies we analysed have high $m_1$ amplitudes, ranging from 2 to 10 percent in the outer parts ($R \gtrsim 3$ kpc), all of which have a physically associated companion. Based on our findings, we advocate the use of the $m=1$ multipole in the mass models used in strong lensing and dynamical studies, particularly for galaxies with recent or ongoing interactions., Comment: 7 pages, 6 figures. Submitted to MNRAS

Self-supervised learning (SSL) is a machine learning approach where the data itself provides supervision, eliminating the need for external labels. The model is forced to learn about the data structure or context by solving a pretext task. With SSL, models can learn from abundant and cheap unlabeled data, significantly reducing the cost of training models where labels are expensive or inaccessible. In Computer Vision, SSL is widely used as pre-training followed by a downstream task, such as supervised transfer, few-shot learning on smaller labeled data sets, and/or unsupervised clustering. Unfortunately, it is infeasible to evaluate SSL methods on all possible downstream tasks and objectively measure the quality of the learned representation. Instead, SSL methods are evaluated using in-domain evaluation protocols, such as fine-tuning, linear probing, and k-nearest neighbors (kNN). However, it is not well understood how well these evaluation protocols estimate the representation quality of a pre-trained model for different downstream tasks under different conditions, such as dataset, metric, and model architecture. We study how classification-based evaluation protocols for SSL correlate and how well they predict downstream performance on different dataset types. Our study includes eleven common image datasets and 26 models that were pre-trained with different SSL methods or have different model backbones. We find that in-domain linear/kNN probing protocols are, on average, the best general predictors for out-of-domain performance. We further investigate the importance of batch normalization and evaluate how robust correlations are for different kinds of dataset domain shifts. We challenge assumptions about the relationship between discriminative and generative self-supervised methods, finding that most of their performance differences can be explained by changes to model backbones.

Satellite galaxies can be used to indicate the dynamical mass of galaxy groups and clusters. In this study, we apply the axis-symmetric Jeans Anisotropic Multi-Gaussian Expansion JAM modeling to satellite galaxies in 28 galaxy clusters selected from the TNG300-1 simulation with halo mass of $\log_{10}M_{200}/M_\odot>14.3$. If using true bound satellites as tracers, the best constrained total mass within the half-mass radius of satellites, $M(

For mass problems $P,Q\subseteq {\mathbb{N}^\mathbb{N}}$ (Baire space), $P$ is Medvedev reducible to $Q$ ($P\leq_sQ$) if for some Turing funcional $\Phi$, $\Phi(Q)\subseteq P$, and Medvedev equivalent to $Q$ if also $Q\leq_sP$. Shafer asked if every closed problem $P$ is Medvedev equivalent to a closed problem $Q$ with $Q\subseteq 2^\mathbb{N}$ (Cantor space). We show that this is not the case., Comment: 10 pages

Recent work has shown that certain integrable and conformal field theories in two dimensions can be given a higher-dimensional origin from holomorphic Chern-Simons in six dimensions. Along with anti-self-dual Yang-Mills and four-dimensional Chern-Simons, this gives rise to a diamond correspondence of theories. In this work we extend this framework to incorporate models realised through gaugings. As well as describing a higher-dimensional origin of coset CFTs, by choosing the details of the reduction from higher dimensions, we obtain rich classes of two-dimensional integrable models including homogeneous sine-Gordon models and generalisations that are new to the literature., Comment: 44 pages