Inference Scheme Research Articles

Constrained cosmological simulations of the Local Group using Bayesian hierarchical field-level inference

We present a novel approach based on Bayesian field-level inference that provides representative initial conditions for simulation of the Local Group (LG) of galaxies and its neighbourhood, constrained by present-day observations. We extended the Bayesian Origin Reconstruction from Galaxies ( algorithm with a multi-resolution approach, allowing us to reach the smaller scales needed to apply the constraints. Our data model simultaneously accounts for observations of mass tracers within the dark haloes of the Milky Way (MW) and M31, for their observed separation and relative velocity, and for the quiet surrounding Hubble flow, represented by the positions and velocities of 31 galaxies at distances between one and four megaparsec. Our approach delivers representative posterior samples of realisations that are statistically and simultaneously consistent with all of these observations, leading to significantly tighter mass constraints than found if the individual datasets are considered separately. In particular, we estimate the virial masses of the MW and M31 to be $ (M_ 200c /M_ and $12.33 respectively, their sum to be $ 200c /M_ and the enclosed mass within spheres of radius $R$ to be $ (M(R)/M_ and $12.96 for $R=1$ Mpc and $3$ Mpc, respectively. The M31-MW orbit is nearly radial for most of our realisations, and most of them feature a dark matter sheet aligning approximately with the supergalactic plane, despite the surrounding density field not being used explicitly as a constraint. High-resolution, high-fidelity resimulations from initial conditions identified using the approximate simulations of our inference scheme continue to satisfy the observational constraints, demonstrating a route to future high-resolution, full-physics simulations of ensembles of LG look-alikes, all of which closely mirror the observed properties of the real system and its immediate environment.

Arithmetic Optimization Algorithm Quantization Optimized With Energy Detection Using Nonparametric Amplitude

ABSTRACTThe spectrum sensing is a major significant task in cognitive radio networks (CRNs) to avoid the unacceptable interference to primary users (PUs). Here, the threshold value determines the effectiveness of spectrum sensing and regarded as a sensing system. The fixed threshold used by the current energy detection‐based spectrum sensing (SS) techniques does not provide sufficient safety for the main users. The threshold is determined by lowering the complete probability of decision error in addition to these guidelines. Therefore, an energy detection using nonparametric amplitude quantization optimized with arithmetic optimization algorithm for enhanced spectrum sensing in CRNs (ED‐NAQ‐AOA‐SS CRN) is proposed in this paper to acquire the ideal threshold for decreasing the total error probability. The proposed method achieves greater probability of detection of 99.67%, 98.38%, 92.34%, and 97.45%, lower settling time of 98.33%, 89.34%, 83.12%, and 88.96%, and lower error rate of 93.15%, 91.25%, 79.90%, and 92.88% compared with existing techniques, like intelligent spectrum sharing and sensing in CRN with adaptive rider optimization algorithm (AROA), a novel technique for spectrum sensing in CRN utilizing fractional gray wolf optimization with the cuckoo search optimization (GWOCS), and adaptive neuro‐fuzzy inference scheme depending on cooperative spectrum sensing optimization in CRNs (ANFIS).

Edge-device collaborative computing for multi-view classification

Motivated by the proliferation of Internet-of-Thing (IoT) devices and the rapid advances in the field of deep learning, there is a growing interest in pushing deep learning computations, conventionally handled by the cloud, to the edge of the network to deliver faster responses to end users, reduce bandwidth consumption to the cloud, and address privacy concerns. However, to fully realize deep learning at the edge, two main challenges still need to be addressed: (i) how to meet the high resource requirements of deep learning on resource-constrained devices, and (ii) how to leverage the availability of multiple streams of spatially correlated data, to increase the effectiveness of deep learning and improve application-level performance. To address the above challenges, we explore collaborative inference at the edge, in which edge nodes and end devices share correlated data and the inference computational burden by leveraging different ways to split computation and fuse data. Besides traditional centralized and distributed schemes for edge-end device collaborative inference, we introduce selective schemes that decrease bandwidth resource consumption by effectively reducing data redundancy. As a reference scenario, we focus on multi-view classification in a networked system in which sensing nodes can capture overlapping fields of view. The proposed schemes are compared in terms of accuracy, computational expenditure at the nodes, communication overhead, inference latency, robustness, and noise sensitivity. Experimental results highlight that selective collaborative schemes can achieve different trade-offs between the above performance metrics, with some of them bringing substantial communication savings (from 18% to 74% of the transmitted data with respect to centralized inference) while still keeping the inference accuracy well above 90%.

Privacy-preserving and verifiable convolution neural network inference and training in cloud computing

With the rapid development of cloud computing, outsourcing massive data and complex deep learning model to cloud servers (CSs) has become a popular trend, which also brings some security problems. One is that the model stored in the CSs may be corrupted, leading to incorrect inference and training results. The other is that the privacy of outsourced data and model may be compromised. However, existing privacy-preserving and verifiable inference schemes suffer from low detection probability, high communication overhead and substantial computational time. To solve the above problems, we propose a privacy-preserving and verifiable scheme for convolutional neural network inference and training in cloud computing. In our scheme, the model owner generates the authenticators for model parameters before uploading the model to CSs. In the phase of model integrity verification, model owner and user can utilize these authenticators to check model integrity with high detection probability. Furthermore, we design a set of privacy-preserving protocols based on replicated secret sharing for both the inference and training phases, significantly reducing communication overhead and computational time. Through security analysis, we demonstrate that our scheme is secure. Experimental evaluations show that the proposed scheme outperforms existing schemes in privacy-preserving inference and model integrity verification.

Exotic Stable Branches with Efficient TOV Sequences

Modern inference schemes for the neutron star (NS) equation of state (EoS) require large numbers of stellar models constructed with different EoS, and these stellar models must capture all the behavior of stable stars. I introduce termination conditions for sequences of stellar models for cold nonrotating NSs that can identify all stable stellar configurations up to arbitrarily large central pressures along with an efficient algorithm to build accurate interpolators for macroscopic properties. I explore the behavior of stars with both high- and low-central pressures. Interestingly, I find that EoSs with monotonically increasing sound-speed can produce multiple stable branches (twin stars) and that large phase transitions at high densities can produce stable branches at nearly any mass scale, including sub-solar masses while still supporting stars with M > 2 M ⊙. I conclude with some speculation about the astrophysical implications of this behavior.

Self-aware collaborative edge inference with embedded devices for IIoT

Edge inference and other compute-intensive industrial Internet of Things (IIoT) applications suffer from a bad quality of experience due to the limited and heterogeneous computing and communication resources of embedded devices. To tackle these issues, we propose a model partitioning-based self-aware collaborative edge inference framework. Specifically, the device can adaptively adjust the local model inference scheme by sensing the available computing and communication resources of surrounding devices. When the inference latency requirement cannot be met by local computation, the model should be partitioned for collaborative computation on other devices to improve the inference efficiency. Furthermore, for two typical IIoT scenarios, i.e., bursting and stacking tasks, the latency-aware and throughput-aware collaborative inference algorithms are designed, respectively. Via jointly optimizing the partition layer and collaborative device selection, the optimal inference efficiency, characterized by minimum inference latency and maximum inference throughput, can be obtained. Finally, the performance of our proposal is validated through extensive simulations and tests conducted on 10 Raspberry Pi 4Bs using popular models. Specifically, in the case of two collaborative devices, our platform reaches up to 92.59% latency reduction for bursting tasks and 16.19× throughput growth for stacking tasks. In addition, the divergence between simulations and tests ranges from 1.64% to 9.56% for bursting tasks and from 3.24% to 11.24% for stacking tasks, which indicates that the theoretical performance analyses are solid. For the general case where the data privacy is not considered and the number of collaborative devices is optimally determined, up to 14.76× throughput speed up and 84.04% latency reduction can be obtained.

An adaptive neuro-fuzzy inference scheme for defect detection and classification of solar PV cells

An adaptive neuro-fuzzy inference scheme for defect detection and classification of solar PV cells

Inferring Kinetics and Entropy Production from Observable Transitions in Partially Accessible, Periodically Driven Markov Networks

For a network of discrete states with a periodically driven Markovian dynamics, we develop an inference scheme for an external observer who has access to some transitions. Based on waiting-time distributions between these transitions, the periodic probabilities of states connected by these observed transitions and their time-dependent transition rates can be inferred. Moreover, the smallest number of hidden transitions between accessible ones and some of their transition rates can be extracted. We prove and conjecture lower bounds on the total entropy production for such periodic stationary states. Even though our techniques are based on generalizations of known methods for steady states, we obtain original results for those as well.

PC-Kriging-powered parallelizing Bayesian updating for stochastic vehicle-track dynamical system with contact force measurements and Gaussian process discrepancy model

High-precision vehicle-track coupled dynamical models play a vital role in assessing the running safety of the vehicle, tracking fatigue behavior, and establishing a digital twin for high-speed railways. This study established a high-fidelity vehicle-track coupled dynamical model as a forward solver, which was then updated in a Bayesian inference framework based on the field tests of rail irregularities and wheel/rail normal contact forces. The prediction errors corresponding to different time steps were represented through a Gaussian process (GP) model characterized by a covariance function with the Gaussian kernel to incorporate the correlation between different time steps. To cope with the considerable expense of the likelihood function calculations with large-scale loop operations and the computational burden due to the large batch of repetitive evaluations of the high-fidelity forward model involved in the stochastic sampling from the posterior probability distribution, a metamodel-powered parallelizing stochastic sampling scheme was adopted. The posterior distribution could be formulated by replacing the expensive explicit model evaluation involved in the likelihood function with a Polynomial-Chaos-Kriging (PC-Kriging) predictor providing a surrogate mapping between the wheel/rail normal contact forces and physical parameters. Finally, a parallelizing computing scheme running across multiple CPU cores on high-performance computers was realized to sample from the vectorized formula of the emulator-powered posterior distribution. Field test data from a high-speed railway in China was adopted to demonstrate that the Bayesian inference scheme can quantify the uncertainties of the core physical parameters of the high-fidelity vehicle-track coupled dynamical system by efficiently trading off accuracy and computational efficiency.

Inference of neutron-star properties with unified crust-core equations of state for parameter estimation

Context. Relating different global neutron-star (NS) properties, such as tidal deformability and radius, or mass and radius, requires an equation of state (EoS). Determining the NS EoS is therefore not only the science goal of a variety of observational projects, but it also enters in the analysis process; for example, to predict a NS radius from a measured tidal deformability via gravitational waves (GW) during the inspiral of a binary NS merger. To this aim, it is important to estimate the theoretical uncertainties on the EoS, one of which is the possible bias coming from an inconsistent treatment of the low-density region; that is, the use of a so called non-unified NS crust. Aims. We propose a numerical tool allowing the user to consistently match a nuclear-physics informed crust to an arbitrary high-density EoS describing the core of the star. Methods. We introduce an inversion procedure of the EoS close to saturation density that allows users to extract nuclear-matter parameters and extend the EoS to lower densities in a consistent way. For the treatment of inhomogeneous matter in the crust, a standard approach based on the compressible liquid-drop (CLD) model approach was used in our work. A Bayesian analysis using a parametric agnostic EoS representation in the high-density region is also presented in order to quantify the uncertainties induced by an inconsistent treatment of the crust. Results. We show that the use of a fixed, realistic-but-inconsistent model for the crust causes small but avoidable errors in the estimation of global NS properties and leads to an underestimation of the uncertainties in the inference of NS properties. Conclusions. Our results highlight the importance of employing a consistent EoS in inference schemes. The numerical tool that we developed to reconstruct such a thermodynamically consistent EoS, CUTER, has been tested and validated for use by the astrophysical community.

Hierarchical segmentation of surgical scenes in laparoscopy.

Segmentation of surgical scenes may provide valuable information for real-time guidance and post-operative analysis. However, in some surgical video frames there is unavoidable ambiguity, leading to incorrect predictions of class or missed detections. In this work, we propose a novel method that alleviates this problem by introducing a hierarchy and associated hierarchical inference scheme that allows broad anatomical structures to be predicted when fine-grained structures cannot be reliably distinguished. First, we formulate a multi-label segmentation loss informed by a hierarchy of anatomical classes and then train a network using this. Subsequently, we use a novel leaf-to-root inference scheme ("Hiera-Mix") to determine the trade-off between label confidence and granularity. This method can be applied to any segmentation model. We evaluate our method using a large laparoscopic cholecystectomy dataset with 65,000 labelled frames. We observed an increase in per-structure detection F1 score for the critical structures, when evaluated across their sub-hierarchies, compared to the baseline method: 6.0% for the cystic artery and 2.9% for the cystic duct, driven primarily by increases in precision of 11.3% and 4.7%, respectively. This corresponded to visibly improved segmentation outputs, with better characterisation of the undissected area containing the critical structures and fewer inter-class confusions. For other anatomical classes, which did not stand to benefit from the hierarchy, performance was unimpaired. Our proposed hierarchical approach improves surgical scene segmentation in frames with ambiguity, by more suitably reflecting the model's parsing of the scene. This may be beneficial in applications of surgical scene segmentation, including recent advancements towards computer-assisted intra-operative guidance.

Estimate of entropy production rate can spatiotemporally resolve the active nature of cell flickering

We use the short-time inference scheme [Manikandan , ], obtained within the framework of stochastic thermodynamics, to infer a lower bound to entropy production rate from flickering data generated by interference reflection microscopy of HeLa cells. We can clearly distinguish active cell membranes from their adenosine-triphosphate-depleted selves and even spatiotemporally resolve activity down to the scale of about 1 µm. Our estimate of activity is . Published by the American Physical Society 2024

Private-preserving language model inference based on secure multi-party computation

With the exponential expansion of Internet information, technology that combines big data and artificial intelligence has gradually developed. Pre-trained large-scale language models with the transformer architecture as the core have begun to be used in daily life, resulting in the huge market of MLaaS. leading to the significant market of Machine Learning as a Service (MLaaS). Although MLaaS brings huge benefits to users, it requires receiving users’ data for processing, which includes many sensitive data. While MLaaS offers considerable benefits, it necessitates processing users’ data, which includes much sensitive data.Therefore, the problem of privacy data leakage has also been exposed. In this article paper, we propose a novel language model secure inference scheme based on secure multi-party computation (MPC) technology. This solution involves three non-colluding parties: the data provider, the model provider, and the computing power provider. Compared with direct inference on pre-trained large models, the proposed security inference framework improves the inference speed by 1.55–6.25 times. Our findings demonstrate that, when compared to conventional inference methods on pre-trained large-scale models, our approach significantly enhances inference efficiency, achieving speed improvements ranging from 1.55 to 6.25 times.

Bayesian uncertainty evaluation applied to the tilted-wave interferometer

The tilted-wave interferometer is a promising technique for the development of a reference measurement system for the highly accurate form measurement of aspheres and freeform surfaces. The technique combines interferometric measurements, acquired with a special setup, and sophisticated mathematical evaluation procedures. To determine the form of the surface under test, a computational model is required that closely mimics the measurement process of the physical measurement instruments. The parameters of the computational model, comprising the surface under test sought, are then tuned by solving an inverse problem. Due to this embedded structure of the real experiment and computational model and the overall complexity, a thorough uncertainty evaluation is challenging. In this work, a Bayesian approach is proposed to tackle the inverse problem, based on a statistical model derived from the computational model of the tilted-wave interferometer. Such a procedure naturally allows for uncertainty quantification to be made. We present an approximate inference scheme to efficiently sample quantities of the posterior using Monte Carlo sampling involving the statistical model. In particular, the methodology derived is applied to the tilted-wave interferometer to obtain an estimate and corresponding uncertainty of the pixel-by-pixel form of the surface under test for two typical surfaces taking into account a number of key influencing factors. A statistical analysis using experimental design is employed to identify main influencing factors and a subsequent analysis confirms the efficacy of the method derived.

TriOS Schwarzschild Orbit Modeling: Robustness of Parameter Inference for Masses and Shapes of Triaxial Galaxies with Supermassive Black Holes

Evidence for the majority of the supermassive black holes in the local Universe has been obtained dynamically from stellar motions with the Schwarzschild orbit superposition method. However, there have been only a handful of studies using simulated data to examine the ability of this method to reliably recover known input black hole masses M BH and other galaxy parameters. Here, we conduct a comprehensive assessment of the reliability of the triaxial Schwarzschild method at simultaneously determining M BH, stellar mass-to-light ratio M*/L, dark matter mass, and three intrinsic triaxial shape parameters of simulated galaxies. For each of 25 rounds of mock observations using simulated stellar kinematics and the TriOS code, we derive best-fitting parameters and confidence intervals after a full search in the 6D parameter space with our likelihood-based model inference scheme. The two key mass parameters, M BH and M*/L, are recovered within the 68% confidence interval, and other parameters are recovered between the 68% and 95% confidence intervals. The spatially varying velocity anisotropy of the stellar orbits is also well recovered. We explore whether the goodness-of-fit measure used for galaxy model selection in our pipeline is biased by variable complexity across the 6D parameter space. In our tests, adding a penalty term to the likelihood measure either makes little difference, or worsens the recovery in some cases.

The Inverse of Exact Renormalization Group Flows as Statistical Inference.

We build on the view of the Exact Renormalization Group (ERG) as an instantiation of Optimal Transport described by a functional convection-diffusion equation. We provide a new information-theoretic perspective for understanding the ERG through the intermediary of Bayesian Statistical Inference. This connection is facilitated by the Dynamical Bayesian Inference scheme, which encodes Bayesian inference in the form of a one-parameter family of probability distributions solving an integro-differential equation derived from Bayes' law. In this note, we demonstrate how the Dynamical Bayesian Inference equation is, itself, equivalent to a diffusion equation, which we dub Bayesian Diffusion. By identifying the features that define Bayesian Diffusion and mapping them onto the features that define the ERG, we obtain a dictionary outlining how renormalization can be understood as the inverse of statistical inference.

Approximate inference for longitudinal mechanistic HIV contact network

Network models are increasingly used to study infectious disease spread. Exponential Random Graph models have a history in this area, with scalable inference methods now available. An alternative approach uses mechanistic network models. Mechanistic network models directly capture individual behaviors, making them suitable for studying sexually transmitted diseases. Combining mechanistic models with Approximate Bayesian Computation allows flexible modeling using domain-specific interaction rules among agents, avoiding network model oversimplifications. These models are ideal for longitudinal settings as they explicitly incorporate network evolution over time. We implemented a discrete-time version of a previously published continuous-time model of evolving contact networks for men who have sex with men and proposed an ABC-based approximate inference scheme for it. As expected, we found that a two-wave longitudinal study design improves the accuracy of inference compared to a cross-sectional design. However, the gains in precision in collecting data twice, up to 18%, depend on the spacing of the two waves and are sensitive to the choice of summary statistics. In addition to methodological developments, our results inform the design of future longitudinal network studies in sexually transmitted diseases, specifically in terms of what data to collect from participants and when to do so.

Short-Time Infrequent Metadynamics for Improved Kinetics Inference.

Infrequent Metadynamics is a popular method to obtain the rates of long time-scale processes from accelerated simulations. The inference procedure is based on rescaling the first-passage times of the Metadynamics trajectories using a bias-dependent acceleration factor. While useful in many cases, it is limited to Poisson kinetics, and a reliable estimation of the unbiased rate requires slow bias deposition and prior knowledge of efficient collective variables. Here, we propose an improved inference scheme, which is based on two key observations: (1) the time-independent rate of Poisson processes can be estimated using short trajectories only. (2) Short trajectories experience minimal bias, and their rescaled first-passage times follow the unbiased distribution even for relatively high deposition rates and suboptimal collective variables. Therefore, by basing the inference procedure on short time scales, we obtain an improved trade-off between speedup and accuracy at no additional computational cost, especially when employing suboptimal collective variables. We demonstrate the improved inference scheme for a model system and two molecular systems.

Double-Bounded Optimal Transport for Advanced Clustering and Classification

Optimal transport (OT) is attracting increasing attention in machine learning. It aims to transport a source distribution to a target one at minimal cost. In its vanilla form, the source and target distributions are predetermined, which contracts to the real-world case involving undetermined targets. In this paper, we propose Doubly Bounded Optimal Transport (DB-OT), which assumes that the target distribution is restricted within two boundaries instead of a fixed one, thus giving more freedom for the transport to find solutions. Based on the entropic regularization of DB-OT, three scaling-based algorithms are devised for calculating the optimal solution. We also show that our DB-OT is helpful for barycenter-based clustering, which can avoid the excessive concentration of samples in a single cluster. Then we further develop DB-OT techniques for long-tailed classification which is an emerging and open problem. We first propose a connection between OT and classification, that is, in the classification task, training involves optimizing the Inverse OT to learn the representations, while testing involves optimizing the OT for predictions. with this OT perspective, we first apply DB-OT to improve the loss, and the Balanced Softmax is shown as a special case. Then we apply DB-OT for inference in the testing process. Even with vanilla Softmax trained features, our experiments show that our method can achieve good results with our improved inference scheme in the testing stage.

Generalized Variational Inference via Optimal Transport

Variational Inference (VI) has gained popularity as a flexible approximate inference scheme for computing posterior distributions in Bayesian models. Original VI methods use Kullback-Leibler (KL) divergence to construct variational objectives. However, KL divergence has zero-forcing behavior and is completely agnostic to the metric of the underlying data distribution, resulting in bad approximations. To alleviate this issue, we propose a new variational objective by using Optimal Transport (OT) distance, which is a metric-aware divergence, to measure the difference between approximate posteriors and priors. The superior performance of OT distance enables us to learn more accurate approximations. We further enhance the objective by gradually including the OT term using a hyperparameter λ for over-parameterized models. We develop a Variational inference method with OT (VOT) which presents a gradient-based black-box framework for solving Bayesian models, even when the density function of approximate distribution is not available. We provide the consistency analysis of approximate posteriors and demonstrate the practical effectiveness on Bayesian neural networks and variational autoencoders.