Inference Strategy Research Articles

Parallax-Tolerant Weakly-Supervised Pixel-Wise Deep Color Correction for Image Stitching of Pinhole Camera Arrays

Camera arrays typically use image-stitching algorithms to generate wide field-of-view panoramas, but parallax and color differences caused by varying viewing angles often result in noticeable artifacts in the stitching result. However, existing solutions can only address specific color difference issues and are ineffective for pinhole images with parallax. To overcome these limitations, we propose a parallax-tolerant weakly supervised pixel-wise deep color correction framework for the image stitching of pinhole camera arrays. The total framework consists of two stages. In the first stage, based on the differences between high-dimensional feature vectors extracted by a convolutional module, a parallax-tolerant color correction network with dynamic loss weights is utilized to adaptively compensate for color differences in overlapping regions. In the second stage, we introduce a gradient-based Markov Random Field inference strategy for correction coefficients of non-overlapping regions to harmonize non-overlapping regions with overlapping regions. Additionally, we innovatively propose an evaluation metric called Color Differences Across the Seam to quantitatively measure the naturalness of transitions across the composition seam. Comparative experiments conducted on popular datasets and authentic images demonstrate that our approach outperforms existing solutions in both qualitative and quantitative evaluations, effectively eliminating visible artifacts and producing natural-looking composite images.

Contrasts between the practice and listening and reading skills teaching of English language

This paper analyses the difference between teaching and practicing listening and reading skills in the context of learning English as a second language (ESL). A systematic literature review examines recent studies published within the last five years on pedagogical approaches. The results indicate that active teaching, which includes keyword identification and inference strategies, is significantly more effective than repetitive practice, allowing students to apply these skills more effectively in real-life situations. It is concluded that structured instruction is key to fostering deep learning, while unguided practice may limit the effective development of these skills.

Countries' progress towards Global Health Security (GHS) increased health systems resilience during the Coronavirus Disease-19 (COVID-19) pandemic: A difference-in-difference study of 191 countries.

Research on health systems resilience during the Coronavirus Disease-2019 pandemic frequently used the Global Health Security Index (GHSI), a composite index scoring countries' health security and related capabilities. Conflicting results raised questions regarding the validity of the GHSI as a reliable index. This study attempted to better characterize when and to what extent countries' progress towards Global Health Security (GHS) augments health systems resilience. We used longitudinal data from 191 countries and a difference-in-difference (DiD) causal inference strategy to quantify the effect of countries' GHS capacity as measured by the GHSI on their coverage rates for essential childhood immunizations, a previously established proxy for health systems resilience. Using a sliding scale of cutoff values with step increments of one, we divided countries into treatment and control groups and determined the lowest GHSI score at which a safeguarding effect was observed. All analyses were adjusted for potential confounders. World Bank governance indicators were employed for robustness tests. While countries with overall GHSI scores of 57 and above prevented declines in childhood immunization coverage rates from 2020-2022 (coef: 0.91; 95% CI: 0.41-1.41), this safeguarding effect was strongest in 2021 (coef: 1.23; 95% CI: 0.05-2.41). Coefficient sizes for overall GHSI scores were smaller compared to several GHSI sub-components, including countries' environmental risks (coef: 4.28; 95% CI: 2.56-5.99) and emergency preparedness and response planning (coef: 1.82; 95% CI: 0.54-3.11). Our findings indicate that GHS was positively associated with health systems resilience during the pandemic (2020) and the following two years (2021-2022), that GHS may have had the most significant protective effects in 2021 as compared with 2020 and 2022, and that countries' underlying characteristics, including governance quality, bolstered health systems resilience during the pandemic.

NVDS +: Towards Efficient and Versatile Neural Stabilizer for Video Depth Estimation.

Video depth estimation aims to infer temporally consistent depth. One approach is to finetune a single-image model on each video with geometry constraints, which proves inefficient and lacks robustness. An alternative is learning to enforce consistency from data, which requires well-designed models and sufficient video depth data. To address both challenges, we introduce NVDS + that stabilizes inconsistent depth estimated by various single-image models in a plug-and-play manner. We also elaborate a large-scale Video Depth in the Wild (VDW) dataset, which contains 14,203 videos with over two million frames, making it the largest natural-scene video depth dataset. Additionally, a bidirectional inference strategy is designed to improve consistency by adaptively fusing forward and backward predictions. We instantiate a model family ranging from small to large scales for different applications. The method is evaluated on VDW dataset and three public benchmarks. To further prove the versatility, we extend NVDS + to video semantic segmentation and several downstream applications like bokeh rendering, novel view synthesis, and 3D reconstruction. Experimental results show that our method achieves significant improvements in consistency, accuracy, and efficiency. Our work serves as a solid baseline and data foundation for learning-based video depth estimation. Code and dataset are available at: https://github.com/RaymondWang987/NVDS.

Role of Design for Causal Analysis in Research on Community Engagement: Review and Recommendations

The purpose of this review is to advance the application of causal inference strategies to service-learning and community engagement and offer recommendations for both practitioners and researchers. The review offers an introduction to various techniques for yielding causal conclusions and discusses an example of the technique from the literature. The conclusion offers recommendations for pursuing causal inference and improving research designs related to service-learning and community engagement.

Parallel implementation and performance of super-resolution generative adversarial network turbulence models for large-eddy simulation

Super-resolution (SR) generative adversarial networks (GANs) are promising for turbulence closure in large-eddy simulation (LES) due to their ability to accurately reconstruct high-resolution data from low-resolution fields. Current model training and inference strategies are not sufficiently mature for large-scale, distributed calculations due to the computational demands and often unstable training of SR-GANs, which limits the exploration of improved model structures, training strategies, and loss-function definitions. Integrating SR-GANs into LES solvers for inference-coupled simulations is also necessary to assess their a posteriori accuracy, stability, and cost. We investigate parallelization strategies for SR-GAN training and inference-coupled LES, focusing on computational performance and reconstruction accuracy. We examine distributed data-parallel training strategies for hybrid CPU–GPU node architectures and the associated influence of low-/high-resolution subbox size, global batch size, and discriminator accuracy. Accurate predictions require training subboxes that are sufficiently large relative to the Kolmogorov length scale. Care should be placed on the coupled effect of training batch size, learning rate, number of training subboxes, and discriminator’s learning capabilities. We introduce a data-parallel SR-GAN training and inference library for heterogeneous architectures that enables exchange between the LES solver and SR-GAN inference at runtime. We investigate the predictive accuracy and computational performance of this arrangement with particular focus on the overlap (halo) size required for accurate SR reconstruction. Similarly, a posteriori parallel scaling for efficient inference-coupled LES is constrained by the SR subdomain size, GPU utilization, and reconstruction accuracy. Based on these findings, we establish guidelines and best practices to optimize resource utilization and parallel acceleration of SR-GAN turbulence model training and inference-coupled LES calculations while maintaining predictive accuracy.

METAPHORICAL INSTANTIATION OF IDIOMATIC EXPRESSIONS IN MODERN POLITICAL DISCOURSE

The present paper is a study of English idiomatic expressions denoting cooperation in terms of Cognitive Linguistic Critical Discourse Analysis, which is herein referred to as a form of applied cognitive linguistics that investigates the links between language, cognition and social action in contexts of political communication. The methodological underpinnings involve principles of Critical Discourse Analysis proper, such as analysis of ideology (a system of values and/or ideas that shape and are shaped by discourse), framing (the way language and discourse shape how individuals or groups interpret and understand certain events, values or ideas), pragmatic inference (understanding intended meaning), lexical choices (words or phrases employed to convey a certain meaning), and discourse strategies (the conscious selection and use of language to achieve specific goals in communication). The cognitive linguistic dimension of the survey relies on the principles of schema induction and hierarchical schematic modeling of the idiomatic expressions under analysis. It is argued that meaning in context, instantiated by an idiomatic expression, is a schematic representation on two distinct conceptual levels: the dynamic, provided for by mental spaces, and the static, realized by frames, conceptual domains, and image schemas. A case study to verify this hypothesis was performed on the idiom stand shoulder to shoulder, and its findings were then employed to exemplify the extended cognitive structure of the idiomatic expressions, such as play ball, build bridges, in concert, hitch horses together, on the same page, close ranks, sing from the same hymn sheet, and it takes two to tango

Inferring strain-level mutational drivers of phage-bacteria interaction phenotypes arising during coevolutionary dynamics.

The enormous diversity of bacteriophages and their bacterial hosts presents a significant challenge to predict which phages infect a focal set of bacteria. Infection is largely determined by complementary-and largely uncharacterized-genetics of adsorption, injection, cell take-over, and lysis. Here we present a machine learning approach to predict phage-bacteria interactions trained on genome sequences of and phenotypic interactions among 51 Escherichia coli strains and 45 phage λ strains that coevolved in laboratory conditions for 37 days. Leveraging multiple inference strategies and without a priori knowledge of driver mutations, this framework predicts both who infects whom and the quantitative levels of infections across a suite of 2,295 potential interactions. We found that the most effective approach inferred interaction phenotypes from independent contributions from phage and bacteria mutations, accurately predicting 86% of interactions while reducing the relative error in the estimated strength of the infection phenotype by 40%. Feature selection revealed key phage λ and Escherchia coli mutations that have a significant influence on the outcome of phage-bacteria interactions, corroborating sites previously known to affect phage λ infections, as well as identifying mutations in genes of unknown function not previously shown to influence bacterial resistance. The method's success in recapitulating strain-level infection outcomes arising during coevolutionary dynamics may also help inform generalized approaches for imputing genetic drivers of interaction phenotypes in complex communities of phage and bacteria.

SecoInfer: Secure DNN End-Edge Collaborative Inference Framework Optimizing Privacy and Latency

End-edge collaborative inference enhances computational efficiency by segmenting a deep neural network (DNN) model into two parts, executed across the end device and the edge node. However, existing collaborative inference strategies often involve transmitting original inputs from the end device to the edge node, resulting in significant risks of user detail leakage without requiring input reconstruction. Therefore, in this work, we present SecoInfer, a secure layer-level DNN end-edge collaborative inference framework. SecoInfer achieves joint optimization of data privacy and inference latency for DNN partition solutions that meet latency constraints, supported by three key designs. First, the privacy-aware DNN layer projection measurement quantifies the difficulty adversaries encounter in reconstructing the original input from the intermediate output of each layer. Then, the latency-privacy integrated structure modeling enables the direct calculation of the privacy measurement and inference latency for each partition solution from a list element or a directed acyclic graph (DAG) cut. Finally, the two-stage latency constraint adjustment scheme narrows down the search space of feasible partition solutions at the block level and fine-tunes the final one to meet the latency constraint based on layer depth. We prototype SecoInfer, utilizing a Raspberry Pi 4B as the end device and a server with an NVIDIA GeForce RTX 3060 GPU as the edge node. Experimental results demonstrate that under latency constraints of 20 ms, 33 ms, and 40 ms, SecoInfer reduces adversarial data reconstruction by 9.84%, 19.26%, and 25.18%, respectively, without any loss of task model accuracy. SecoInfer also enhances efficiency, reducing the time needed to determine optimal end-edge partition solutions on a Raspberry Pi 4B by 18.04%.

Inferring strain-level mutational drivers of phage-bacteria interaction phenotypes arising during coevolutionary dynamics.

The enormous diversity of bacteriophages and their bacterial hosts presents a significant challenge to predict which phages infect a focal set of bacteria. Infection is largely determined by complementary - and largely uncharacterized - genetics of adsorption, injection, cell take-over and lysis. Here we present a machine learning approach to predict phage-bacteria interactions trained on genome sequences of and phenotypic interactions amongst 51 Escherichia coli strains and 45 phage strains that coevolved in laboratory conditions for 37 days. Leveraging multiple inference strategies and without a priori knowledge of driver mutations, this framework predicts both who infects whom and the quantitative levels of infections across a suite of 2,295 potential interactions. We found that the most effective approach inferred interaction phenotypes from independent contributions from phage and bacteria mutations, accurately predicting of interactions while reducing the relative error in the estimated strength of the infection phenotype by . Feature selection revealed key phage and E. coli mutations that have a significant influence on the outcome of phage-bacteria interactions, corroborating sites previously known to affect phage infections, as well as identifying mutations in genes of unknown function not previously shown to influence bacterial resistance. The method's success in recapitulating strain-level infection outcomes arising during coevolutionary dynamics may also help inform generalized approaches for imputing genetic drivers of interaction phenotypes in complex communities of phage and bacteria.

Adaptive human–machine theorem proving system

This paper presents a novel approach to constructing human–machine theorem proving systems. These systems integrate machine learning capabilities, human expert knowledge, and rigorous logical control for the effective construction and verification of proofs. The innovation of this approach lies in its openness: the user is given a tool for building such systems. Users can create theorem proving systems by selecting existing logical inference strategies or adding new ones in accordance with the provided interfaces or relationship agreements. The systems being built have a significant human--machine adaptive nature. During their operation, a meta-strategy is developed and trained based on the foundational elements of the existing strategies. The system is designed as a universal framework for managing various strategies, with the provision of a basic architecture and a library of strategies with the possibility of adding new ones. During the learning process, a system of structural characteristics is also accumulated and trained, on the basis of which a decision is made on the use of specific strategy elements at the next step. The presentation is conducted for the sequential calculus of minimal positive predicate logic as the most suitable for the deductive synthesis of programs and algorithms, for which it is proposed to use this approach.

Fusing semantic and syntactic information for aspect sentiment triplet extraction

Aspect Sentiment Triplet Extraction (ASTE) aims to extract aspect terms, sentiment polarity and opinion terms explaining the reason for the sentiment from a sentence in the form of triplets. Many existing studies model the context by graph neural networks to learn relevant information from the generated graphs. However, some sentences may have syntactic errors or lack significant grammar, which may lead to poor results on the dataset of the model. In this paper, we propose the Fusing Semantic and Syntactic Information for Aspect Sentiment Triplet Extraction (FSSI) model, which incorporates both syntactic structure and semantic relevance in the context. Specifically, we construct a syntactic graph convolutional network to obtain comprehensive syntactic structure information and a semantic graph convolutional network to obtain global semantic relevance of sentences using the self-attention mechanism. Furthermore, we concatenate the graph representations generated by the two graph convolution networks to obtain the final enhanced representation. Finally, we apply an effective inference strategy to extract triplets. Extensive experimental results on benchmark datasets show that our model outperforms state-of-the-art approaches.

Active learning strategies for the design of sustainable alloys.

Active learning comprises machine learning-based approaches that integrate surrogate model inference, exploitation and exploration strategies with active experimental feedback into a closed-loop framework. This approach aims at describing and predicting specific material properties, without requiring lengthy, expensive or repetitive experiments. Recently, active learning has shown potential as an approach for the design of sustainable materials, such as scrap-compatible alloys, and for enhancing the longevity of metallic materials. However, in-depth investigations into suited best-practice strategies of active learning for sustainable materials science are still scarce. This study aims to present and discuss active learning strategies for developing and improving sustainable alloys, addressing single-objective and multi-objective learning and modelling scenarios. As model cases, we discuss active learning strategies for optimizing Invar and magnetic alloys, representing single-objective scenarios, and more general steel design approaches, exemplifying multi-objective optimization. We discuss the significance of finding the right balance between exploitation and exploration strategies in active learning and suggest strategies to reduce the number of iterations across diverse scenarios. This kind of research aims to find metrics for a more effective application of active learning and is used here to advance the field of sustainable alloy design.This article is part of the discussion meeting issue 'Sustainable metals: science and systems'.

Multi-Channel Optimization Generative Model for Stable Ultra-Sparse-View CT Reconstruction.

Score-based generative model (SGM) has risen to prominence in sparse-view CT reconstruction due to its impressive generation capability. The consistency of data is crucial in guiding the reconstruction process in SGM-based reconstruction methods. However, the existing data consistency policy exhibits certain limitations. Firstly, it employs partial data from the reconstructed image of the iteration process for image updates, which leads to secondary artifacts with compromising image quality. Moreover, the updates to the SGM and data consistency are considered as distinct stages, disregarding their interdependent relationship. Additionally, the reference image used to compute gradients in the reconstruction process is derived from the intermediate result rather than ground truth. Motivated by the fact that a typical SGM yields distinct outcomes with different random noise inputs, we propose a Multi-channel Optimization Generative Model (MOGM) for stable ultra-sparse-view CT reconstruction by integrating a novel data consistency term into the stochastic differential equation model. Notably, the unique aspect of this data consistency component is its exclusive reliance on original data for effectively confining generation outcomes. Furthermore, we pioneer an inference strategy that traces back from the current iteration result to ground truth, enhancing reconstruction stability through foundational theoretical support. We also establish a multi-channel optimization reconstruction framework, where conventional iterative techniques are employed to seek the reconstruction solution. Quantitative and qualitative assessments on 23 views datasets from numerical simulation, clinical cardiac and sheep's lung underscore the superiority of MOGM over alternative methods. Reconstructing from just 10 and 7 views, our method consistently demonstrates exceptional performance.

Scanning sample-specific miRNA regulation from bulk and single-cell RNA-sequencing data

BackgroundRNA-sequencing technology provides an effective tool for understanding miRNA regulation in complex human diseases, including cancers. A large number of computational methods have been developed to make use of bulk and single-cell RNA-sequencing data to identify miRNA regulations at the resolution of multiple samples (i.e. group of cells or tissues). However, due to the heterogeneity of individual samples, there is a strong need to infer miRNA regulation specific to individual samples to uncover miRNA regulation at the single-sample resolution level.ResultsHere, we develop a framework, Scan, for scanning sample-specific miRNA regulation. Since a single network inference method or strategy cannot perform well for all types of new data, Scan incorporates 27 network inference methods and two strategies to infer tissue-specific or cell-specific miRNA regulation from bulk or single-cell RNA-sequencing data. Results on bulk and single-cell RNA-sequencing data demonstrate the effectiveness of Scan in inferring sample-specific miRNA regulation. Moreover, we have found that incorporating the prior information of miRNA targets can generally improve the accuracy of miRNA target prediction. In addition, Scan can contribute to construct cell/tissue correlation networks and recover aggregate miRNA regulatory networks. Finally, the comparison results have shown that the performance of network inference methods is likely to be data-specific, and selecting optimal network inference methods is required for more accurate prediction of miRNA targets.ConclusionsScan provides a useful method to help infer sample-specific miRNA regulation for new data, benchmark new network inference methods and deepen the understanding of miRNA regulation at the resolution of individual samples.

Double-Layer Distributed and Integrated Fault Detection Strategy for Non-Gaussian Dynamic Industrial Systems.

Currently, with the increasing scale of industrial systems, multisensor monitoring data exhibit large-scale dynamic Gaussian and non-Gaussian concurrent complex characteristics. However, the traditional principal component analysis method is based on Gaussian distribution and uncorrelated assumptions, which are greatly limited in practice. Therefore, developing a new fault detection method for large-scale Gaussian and non-Gaussian concurrent dynamic systems is one of the urgent challenges to be addressed. To this end, a double-layer distributed and integrated data-driven strategy based on Laplacian score weighting and integrated Bayesian inference is proposed. Specifically, in the first layer of the distributed strategy, we design a Jarque-Bera test module to divide all multisensor monitoring variables into Gaussian and non-Gaussian blocks, successfully solving the problem of different data distributions. In the second layer of the distributed strategy, we design a dynamic augmentation module to solve dynamic problems, a K-means clustering module to mine local similarity information of variables, and a Laplace scoring module to quantitatively evaluate the structural retention ability of variables. Therefore, this double-layer distributed strategy can simultaneously combine the different distribution characteristics, dynamism, local similarity, and importance of variables, comprehensively mining the local information of the multisensor data. In addition, we develop an integrated Bayesian inference strategy based on detection performance weighting, which can emphasize the differential contribution of local models. Finally, the fault detection results for the Tennessee Eastman production system and a diesel engine working system validate the superiority of the proposed method.

Bridging the gap between object detection in close-up and high-resolution wide shots

Recent years have seen a significant rise in gigapixel-level image/video capture systems and benchmarks with high-resolution wide (HRW) shots. Different from close-up shots like MS COCO, the higher resolution and wider field of view raise new research and application problems, such as how to perform accurate and efficient object detection with such large input in low-power edge devices like UAVs. There are several unique challenges in HRW shots. (1) Sparse information: the objects of interest cover less area. (2) Various scale: there is 10 to 100× object scale change in one single image. (3) Incomplete objects: the sliding window strategy to handle the large input leads to truncated objects at the window edge. (4) Multi-scale information: it is unclear how to use multi-scale information in training and inference. Consequently, directly using a close-up detector leads to inaccuracy and inefficiency. In this paper, we systematically investigate this problem and bridge the gap between object detection in close-up and HRW shots, by introducing a novel sparse architecture that can be integrated with common networks like ConvNet and Transformer. It leverages alternative sparse learning to complementarily fuse coarse-grained and fine-grained features to (1) adaptively extract valuable information from (2) different object scales. We also propose a novel Cross-window Non-Maximum Suppression (C-NMS) algorithm to (3) improve the box merge from different windows. Furthermore, we propose a (4) simple yet effective multi-scale training and inference strategy to improve accuracy. Experiments on two benchmarks with HRW shots, PANDA and DOTA-v1.0, demonstrate that our methods significantly improve accuracy (up to 5.8%) and speed (up to 3×) over SotAs, for both ConvNet or Transformer based detectors, on edge devices. Our code is open-sourced and available at https://github.com/liwenxi/SparseFormer.

Simulation-based Inference of Reionization Parameters from 3D Tomographic 21 cm Light-cone Images. II. Application of Solid Harmonic Wavelet Scattering Transform

The information regarding how the intergalactic medium is reionized by astrophysical sources is contained in the tomographic three-dimensional 21 cm images from the epoch of reionization. In Zhao et al. (“Paper I”), we demonstrated for the first time that density estimation likelihood-free inference (DELFI) can be applied efficiently to perform a Bayesian inference of the reionization parameters from the 21 cm images. Nevertheless, the 3D image data needs to be compressed into informative summaries as the input of DELFI by, e.g., a trained 3D convolutional neural network (CNN) as in Paper I (DELFI-3D CNN). Here in this paper, we introduce an alternative data compressor, the solid harmonic wavelet scattering transform (WST), which has a similar, yet fixed (i.e., no training), architecture to CNN, but we show that this approach (i.e., solid harmonic WST with DELFI) outperforms earlier analyses based on 3D 21 cm images using DELFI-3D CNN in terms of credible regions of parameters. Realistic effects, including thermal noise and residual foreground after removal, are also applied to the mock observations from the Square Kilometre Array. We show that under the same inference strategy using DELFI, the 21 cm image analysis with solid harmonic WST outperforms the 21 cm power spectrum analysis. This research serves as a proof of concept, demonstrating the potential to harness the strengths of WST and simulation-based inference to derive insights from future 21 cm light-cone image data.

Monte Carlo inference for semiparametric Bayesian regression

Data transformations are essential for broad applicability of parametric regression models. However, for Bayesian analysis, joint inference of the transformation and model parameters typically involves restrictive parametric transformations or nonparametric representations that are computationally inefficient and cumbersome for implementation and theoretical analysis, which limits their usability in practice. This paper introduces a simple, general, and efficient strategy for joint posterior inference of an unknown transformation and all regression model parameters. The proposed approach directly targets the posterior distribution of the transformation by linking it with the marginal distributions of the independent and dependent variables, and then deploys a Bayesian nonparametric model via the Bayesian bootstrap. Crucially, this approach delivers (1) joint posterior consistency under general conditions, including multiple model misspecifications, and (2) efficient Monte Carlo (not Markov chain Monte Carlo) inference for the transformation and all parameters for important special cases. These tools apply across a variety of data domains, including real-valued, positive, and compactly-supported data. Simulation studies and an empirical application demonstrate the effectiveness and efficiency of this strategy for semiparametric Bayesian analysis with linear models, quantile regression, and Gaussian processes. The R package SeBR is available on CRAN.

Disentangling the effects of similarity, familiarity, and liking on social inference strategies.

People constantly make inferences about others' beliefs and preferences. People can draw on various sources of information to make these inferences, including stereotypes, self-knowledge, and target-specific knowledge. What leads people to use each of these sources of information over others? The current study examined factors that influence the use of these sources of information, focusing on three interpersonal dimensions - the extent to which people feel (a) familiar with, (b) similar to, or (c) like the target. In four studies (total N = 1136), participants inferred the beliefs and preferences of others - celebrities (Studies 1a-1b), constructed fictional targets (Study 2), and actual acquaintances (Study 3). Participants also rated familiarity with, similarity to, and liking of the target. Analyses assessed the use of each source of information by comparing inferences with information provided by those sources. Familiarity was associated with greater use of target-specific knowledge, while similarity and liking were associated with self-knowledge. Low similarity and high liking were associated with increased use of stereotypes. We discuss the implication of these findings and their applicability to unique cases, including inferences about celebrities, public figures, and positively stereotyped groups, in which familiarity, similarity, and liking do not perfectly align.