Fine-grained Properties Research Articles

Alternating minimization for generalized rank-1 matrix sensing: sharp predictions from a random initialization

Abstract We consider the problem of estimating the factors of a rank-$1$ matrix with i.i.d. Gaussian, rank-$1$ measurements that are nonlinearly transformed and corrupted by noise. Considering two prototypical choices for the nonlinearity, we study the convergence properties of a natural alternating update rule for this non-convex optimization problem starting from a random initialization. We show sharp convergence guarantees for a sample-split version of the algorithm by deriving a deterministic one-step recursion that is accurate even in high-dimensional problems. Notably, while the infinite-sample population update is uninformative and suggests exact recovery in a single step, the algorithm—and our deterministic one-step prediction—converges geometrically fast from a random initialization. Our sharp, non-asymptotic analysis also exposes several other fine-grained properties of this problem, including how the nonlinearity and noise level affect convergence behaviour. On a technical level, our results are enabled by showing that the empirical error recursion can be predicted by our deterministic one-step updates within fluctuations of the order $n^{-1/2}$ when each iteration is run with $n$ observations. Our technique leverages leave-one-out tools originating in the literature on high-dimensional $M$-estimation and provides an avenue for sharply analyzing complex iterative algorithms from a random initialization in other high-dimensional optimization problems with random data.

Evaluating Large Language Models for Structured Science Summarization in the Open Research Knowledge Graph

Structured science summaries or research contributions using properties or dimensions beyond traditional keywords enhance science findability. Current methods, such as those used by the Open Research Knowledge Graph (ORKG), involve manually curating properties to describe research papers’ contributions in a structured manner, but this is labor-intensive and inconsistent among human domain-expert curators. We propose using Large Language Models (LLMs) to automatically suggest these properties. However, it is essential to assess the readiness of LLMs like GPT-3.5, Llama 2, and Mistral for this task before their application. Our study performs a comprehensive comparative analysis between the ORKG’s manually curated properties and those generated by the aforementioned state-of-the-art LLMs. We evaluate LLM performance from four unique perspectives: semantic alignment with and deviation from ORKG properties, fine-grained property mapping accuracy, SciNCL embedding-based cosine similarity, and expert surveys comparing manual annotations with LLM outputs. These evaluations occur within a multidisciplinary science setting. Overall, LLMs show potential as recommendation systems for structuring science, but further fine-tuning is recommended to improve their alignment with scientific tasks and mimicry of human expertise.

Utilizing N-hydroxy-9-octadecenamide as a collector in flotation separation of bastnaesite and fluorite

Bastnaesite ((Ce,La,Pr,Nd)CO3F) is a significant light rare earth mineral found in nature, known for its fine-grained properties. Flotation is commonly employed for the recovery of fine-grained bastnaesite particles. Collectors serve as an essential flotation reagent that enhance the surface hydrophobicity of target minerals. A novel collector, N-hydroxy-9-octadecenamide (N-OH-9-ODA), was synthesised in this study. N-OH-9-ODA exhibits superior selectivity compared to the traditional collector oleic acid in the flotation separation of bastnaesite and fluorite. The experimental and computational results indicate that N-OH-9-ODA exhibits superior selectivity due to its higher adsorption affinity for bastnaesite surface compared to fluorite surface. The zeta potential and the binding energies of the Ce 3d peaks in the X-ray photoelectron spectrum (XPS) of bastnaesite surface exhibit significant shifts. Conversely, fluorite surface demonstrates minimal alterations in its zeta potential and the binding energies of the Ca 2p peaks in its XPS after its interaction with N-OH-9-ODA.

Maestro: Automatic Generation of Comprehensive Benchmarks for Question Answering Over Knowledge Graphs

Recently, there has been an upsurge in the number of knowledge graphs (KG) that can only be accessed by experts. Non-expert users lack an adequate understanding of the queried knowledge graph's vocabulary and structure, as well as the syntax of the structured query language used to express the user's information needs. To increase the user base of these KGs, a set of Question Answering (QA) systems that use natural language to query these knowledge graphs have been introduced. However, finding a benchmark that accurately evaluates the quality of a QA system is a difficult task due to (1) the high degree of variation in the fine-grained properties among the existing benchmarks, (2) the static nature of the existing benchmarks versus the evolving nature of KGs, and (3) the limited number of KGs targeted by existing benchmarks, which hinders the usability of QA systems in real-world deployment over KGs that are different from those that were used in the evaluation of the QA systems. In this paper, we introduce Maestro, a benchmark generation system for question answering over knowledge graphs. Maestro can generate a new benchmark for any KG given the KG and, optionally, a text corpus that covers this KG. The benchmark generated by Maestro is guaranteed to cover all the properties of the natural language questions and queries that were encountered in the literature as long as the targeted KG includes these properties. Maestro also generates high-quality natural language questions with various utterances that are on par with manually-generated ones to better evaluate QA systems.

Probing sign structure using measurement-induced entanglement

The sign structure of quantum states is closely connected to quantum phases of matter, yet detecting such fine-grained properties of amplitudes is subtle. Here we employ as a diagnostic measurement-induced entanglement (MIE): the average entanglement generated between two parties after measuring the rest of the system. We propose that for a sign-free state, the MIE upon measuring in the sign-free basis decays no slower than correlations in the state before measurement. Concretely, we prove that MIE is upper bounded by mutual information for sign-free stabilizer states (essentially CSS codes), which establishes a bound between scaling dimensions of conformal field theories describing measurement-induced critical points in stabilizer systems. We also show that for sign-free qubit wavefunctions, MIE between two qubits is upper bounded by a simple two-point correlation function, and we verify our proposal in several critical ground states of one-dimensional systems, including the transverse field and tri-critical Ising models. In contrast, for states with sign structure, such bounds can be violated, as we illustrate in critical hybrid circuits involving both Haar or Clifford random unitaries and measurements, and gapless symmetry-protected topological states.

SmartBench

In recent years, a significant number of question answering (QA) systems that retrieve answers to natural language questions from knowledge graphs (KG) have been introduced. However, finding a benchmark that accurately evaluates the quality of a question answering system is a difficult task because of (1) the high degree of variations with respect to the fine-grained properties among the available benchmarks, (2) the static nature of the available benchmarks versus the evolving nature of KGs, and (3) the limited number of KGs targeted by existing benchmarks, which hinders the usability of QA systems in real deployment over KGs that are different from those which the QA system was evaluated using. In this demonstration, we introduce SmartBench, an automatic benchmark generating system for QA over any KG. The benchmark generated by SmartBench is guaranteed to cover all the properties of the natural language questions and queries that were encountered in the literature as long as the targeted KG includes these properties.

PABAFT: Channel Prediction Approach Based on Autoregression and Flexible TDD for 5G Systems

To achieve high gains from multiple antennas in 5G systems, the base station (gNB) constructs precoders using channel measurements obtained based on pilot signals. For high user mobility, the measurements quickly become outdated, which is especially crucial for ultra-reliable low latency communications (URLLC) traffic because it increases channel resource consumption to provide highly reliable transmissions and, consequently, reduces system capacity. Frequent pilot transmissions can provide accurate channel estimation and high-quality precoder but lead to huge overhead. Fortunately, 5G systems enable flexible time division duplex (TDD), which allows the gNB to dynamically change the configuration of downlink and uplink slots and tune the period of channel measurements. The paper exploits this feature and designs a new prediction approach based on autoregression and flexible TDD (PABAFT) that forecasts the channel between consequent pilots transmissions. To learn fine-grained channel properties, the gNB configures uplink pilot transmission in each slot. When the training data are collected, and the model is fitted, the gNB switches back to the regular slot configuration with a long pilot transmission period. Extensive simulations with NS-3 in high-mobility scenarios show that PABAFT provides the signal-to-noise ratio (SNR) close to that with the ideal knowledge of the channel at the gNB. In addition, PABAFT significantly reduces channel resource consumption and, thus, increases capacity for URLLC traffic in comparison to the existing solutions.

Error Bounds for Lanczos-Based Matrix Function Approximation

We analyze the Lanczos method for matrix function approximation (Lanczos-FA), an iterative algorithm for computing $f(\mathbf{A}) \mathbf{b}$ when $\mathbf{A}$ is a Hermitian matrix and $\mathbf{b}$ is a given vector. Assuming that $f : \mathbb{C} \rightarrow \mathbb{C}$ is piecewise analytic, we give a framework, based on the Cauchy integral formula, which can be used to derive a priori and a posteriori error bounds for Lanczos-FA in terms of the error of Lanczos used to solve linear systems. Unlike many error bounds for Lanczos-FA, these bounds account for fine-grained properties of the spectrum of $\mathbf{A}$, such as clustered or isolated eigenvalues. Our results are derived assuming exact arithmetic, but we show that they are easily extended to finite precision computations using existing theory about the Lanczos algorithm in finite precision. We also provide generalized bounds for the Lanczos method used to approximate quadratic forms $\mathbf{b}^\textsf{H} f(\mathbf{A}) \mathbf{b}$, and demonstrate the effectiveness of our bounds with numerical experiments.

An adaptive access control scheme based on trust degrees for edge computing

Access control is an essential mechanism in collaborative environments, which guarantees the security of shared resources. With the intertwinement of emerging technologies likes edge computing, cloud, Artificial Intelligence (AI), and Internet of Things (IoT), access control will encounter its innovation and development shortly. Whereas, traditional access control did not consider the adaptive and fine-grained requirements in edge computing. In addition, the integrity of the accessed data is not considered in the traditional access control scheme. In this paper, inspired by the technologies in AI and cloud, an adaptive access control scheme based on trust degrees is proposed. The trust degrees of users and the quality of data are defined. On this basis, the trust degrees-based adaptive access control framework is proposed. Moreover, in the proposed scheme, technologies in AI and cryptography are converged. In particular, the ciphertext policy attribute-based encryption (CP-ABE) is integrated with the classification and recommendation algorithms in AI, thereby providing adaptive and fine-grained properties for access control in edge computing. Moreover, the integrity of the accessed data is guaranteed by the auditing technique in cryptography. Finally, the proposed access control scheme is implemented by using the PBC library. For supporting the efficient computation in edge computing, all types of elliptic curves provided by the PBC library are tested. The experimental result shows that Type D curve is the most efficient in implementing the proposed scheme.

Person Re-Identification With Hierarchical Discriminative Spatial Aggregation

Practically, person re-identification (re-ID) may suffer from the critical spatial misalignment problem due to inaccurate human detection, variation on human pose and camera viewpoint, etc. To address this, a hierarchical discriminative spatial aggregation method is proposed. The key idea is to conduct spatial aggregation on local human parts via global average-pooling to acquire the strong spatial misalignment tolerance, with VALD encoding on the local parts for facilitating discriminative power jointly. This proposition is built on NetVLAD to ensure end-to-end deep learning capacity. Due to the fine-grained property of person re-ID task that has not been well concerned by the original NetVLAD model for scene recognition, a feature refinement layer that consists of 1 fully-connected (FC) layer and 2 batch normalization (BN) layers is added on top of the raw NetVLAD layer to enhance the discriminative power and training convergence. And, a human body occlusion and background component dropout manner is also proposed to resist the effect of serious occlusion. Technically, a refined codeword initialization manner is proposed to alleviate the potential codeword imbalance problem caused by naive random initialization. The proposed discriminative spatial aggregation approach is then conducted on multi-resolution convolutional feature map layers hierarchically via early feature fusion, to involve richer semantic and fine-grained visual clues jointly. Wide-range experiments on 6 datasets (i.e., CUHK03, DukeMTMC-reID, Occluded-DukeMTMC, Market-1501, MSMT17 and Occluded-REID) verifies the effectiveness of our proposition. The source code and supporting material is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/zmyme/HDSA-reID</uri> .

Ontology-Enabled Emotional Sentiment Analysis on COVID-19 Pandemic-Related Twitter Streams.

The exponential growth of social media users has changed the dynamics of retrieving the potential information from user-generated content and transformed the paradigm of information-retrieval mechanism with the novel developments on the concept of “web of data”. In this regard, our proposed Ontology-Based Sentiment Analysis provides two novel approaches: First, the emotion extraction on tweets related to COVID-19 is carried out by a well-formed taxonomy that comprises possible emotional concepts with fine-grained properties and polarized values. Second, the potential entities present in the tweet can be analyzed for semantic associativity. The extraction of emotions can be performed in two cases: (i) words directly associated with the emotional concepts present in the taxonomy and (ii) words indirectly present in the emotional concepts. Though the latter case is very challenging in processing the tweets to find the hidden patterns and extract the meaningful facts associated with it, our proposed work is able to extract and detect almost 81% of true positives and considerably able to detect the false negatives. Finally, the proposed approach's superior performance is witnessed from its comparison with other peer-level approaches.

CBench

A plethora of question answering (QA) systems that retrieve answers to natural language questions from knowledge graphs have been developed in recent years. However, choosing a benchmark to accurately assess the quality of a question answering system is a challenging task due to the high degree of variations among the available benchmarks with respect to their fine-grained properties. In this demonstration, we introduce CBench, an extensible, and more informative benchmarking suite for analyzing benchmarks and evaluating QA systems. CBench can be used to analyze existing benchmarks with respect to several fine-grained linguistic, syntactic, and structural properties of the questions and queries in the benchmarks. Moreover, CBench can be used to facilitate the evaluation of QA systems using a set of popular benchmarks that can be augmented with other user-provided benchmarks. CBench not only evaluates a QA system based on popular single-number metrics but also gives a detailed analysis of the linguistic, syntactic, and structural properties of answered and unanswered questions to help the developers of QA systems to better understand where their system excels and where it struggles.

Diabetic Retinal Grading Using Attention-Based Bilinear Convolutional Neural Network and Complement Cross Entropy.

Diabetic retinopathy (DR) is a common complication of diabetes mellitus (DM), and it is necessary to diagnose DR in the early stages of treatment. With the rapid development of convolutional neural networks in the field of image processing, deep learning methods have achieved great success in the field of medical image processing. Various medical lesion detection systems have been proposed to detect fundus lesions. At present, in the image classification process of diabetic retinopathy, the fine-grained properties of the diseased image are ignored and most of the retinopathy image data sets have serious uneven distribution problems, which limits the ability of the network to predict the classification of lesions to a large extent. We propose a new non-homologous bilinear pooling convolutional neural network model and combine it with the attention mechanism to further improve the network’s ability to extract specific features of the image. The experimental results show that, compared with the most popular fundus image classification models, the network model we proposed can greatly improve the prediction accuracy of the network while maintaining computational efficiency.

Using automated acoustic analysis to explore the link between planning and articulation in second language speech production

ABSTRACT Speakers learning a second language show systematic differences from native speakers in the retrieval, planning, and articulation of speech. A key challenge in examining the interrelationship between these differences at various stages of production is the need for manual annotation of fine-grained properties of speech. We introduce a new method for automatically analysing voice onset time (VOT), a key phonetic feature indexing differences in sound systems cross-linguistically. In contrast to previous approaches, our method allows reliable measurement of prevoicing, a dimension of VOT variation used by many languages. Analysis of VOTs, word durations, and reaction times from German-speaking learners of Spanish (Baus et al., 2013) suggest that while there are links between the factors impacting planning and articulation, these two processes also exhibit some degree of independence. We discuss the implications of these findings for theories of speech production and future research in bilingual language processing.

Influence of “MC-Bauchemie” Additions on the Fine-Grained Concrete Properties

The paper is devoted to research of influence of “MC-Bauchemie” additions on the fine-grained concrete properties, namely compressive strength. The results of testing of fine-grained concrete made on the basis of two different natural sands are presented. Characteristics of the initial materials used in the work (Portland cement, sands and additions-plasticizers) are given. The basic methods of preparation of mixes and testing of hardened cement stone and concrete are described. The optimal quantity of the selected additions on the example of cement paste and stone are established. Mobility dependences of concrete mixes on a type and quantity of plasticizers are received. The dependence of the additions effectiveness on the mobility of the concrete mix on the quality of the fine aggregate is shown. It is established that the most effective additive is PowerFlow 3100, its optimal amount for obtaining fine-grained concrete on the basis of JSC “Kar’er Myisyi” sand is 0.2 %.

CoCon: A Conference Management System with Formally Verified Document Confidentiality

We present a case study in formally verified security for realistic systems: the information flow security verification of the functional kernel of a web application, the CoCon conference management system. We use the Isabelle theorem prover to specify and verify fine-grained confidentiality properties, as well as complementary safety and “traceback” properties. The challenges posed by this development in terms of expressiveness have led to bounded-deducibility security, a novel security model and verification method generally applicable to systems describable as input/output automata.

Deductive verification with ghost monitors

We present a new approach to deductive program verification based on auxiliary programs called ghost monitors . This technique is useful when the syntactic structure of the target program is not well suited for verification, for example, when an essentially recursive algorithm is implemented in an iterative fashion. Our approach consists in implementing, specifying, and verifying an auxiliary program that monitors the execution of the target program, in such a way that the correctness of the monitor entails the correctness of the target. The ghost monitor maintains the necessary data and invariants to facilitate the proof. It can be implemented and verified in any suitable framework, which does not have to be related to the language of the target programs. This technique is also applicable when we want to establish relational properties between two target programs written in different languages and having different syntactic structure. We then show how ghost monitors can be used to specify and prove fine-grained properties about the infinite behaviors of target programs. Since this cannot be easily done using existing verification frameworks, we introduce a dedicated language for ghost monitors, with an original construction to catch and handle divergent executions. The soundness of the underlying program logic is established using a particular flavor of transfinite games. This language and its soundness are formalized and mechanically checked.

The semantics of emotion in false memory.

The emotional valence of target information has been a centerpiece of recent false memory research, but in most experiments, it has been confounded with emotional arousal. We sought to clarify the results of such research by identifying a shared mathematical relation between valence and arousal ratings in commonly administered normed materials. That relation was then used to (a) decide whether arousal as well as valence influences false memory when they are confounded and to (b) determine whether semantic properties that are known to affect false memory covary with valence and arousal ratings. In Study 1, we identified a quadratic relation between valence and arousal ratings of words and pictures that has 2 key properties: Arousal increases more rapidly as function of negative valence than positive valence, and hence, a given level of negative valence is more arousing than the same level of positive valence. This quadratic function predicts that if arousal as well as valence affects false memory when they are confounded, false memory data must have certain fine-grained properties. In Study 2, those properties were absent from norming data for the Cornell-Cortland Emotional Word Lists, indicating that valence but not arousal affects false memory in those norms. In Study 3, we tested fuzzy-trace theory's explanation of that pattern: that valence ratings are positively related to semantic properties that are known to increase false memory, but arousal ratings are not. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

Mapping binary copolymer property space with neural networks.

The extremely large number of unique polymer compositions that can be achieved through copolymerisation makes it an attractive strategy for tuning their optoelectronic properties. However, this same attribute also makes it challenging to explore the resulting property space and understand the range of properties that can be realised. In an effort to enable the rapid exploration of this space in the case of binary copolymers, we train a neural network using a tiered data generation strategy to accurately predict the optical and electronic properties of 350 000 binary copolymers that are, in principle, synthesizable from their dihalogen monomers via Yamamoto, or Suzuki-Miyaura and Stille coupling after one-step functionalisation. By extracting general features of this property space that would otherwise be obscured in smaller datasets, we identify simple models that effectively relate the properties of these copolymers to the homopolymers of their constituent monomers, and challenge common ideas behind copolymer design. We find that binary copolymerisation does not appear to allow access to regions of the optoelectronic property space that are not already sampled by the homopolymers, although it conceptually allows for more fine-grained property control. Using the large volume of data available, we test the hypothesis that copolymerisation of 'donor' and 'acceptor' monomers can result in copolymers with a lower optical gap than their related homopolymers. Overall, despite the prevalence of this concept in the literature, we observe that this phenomenon is relatively rare, and propose conditions that greatly enhance the likelihood of its experimental realisation. Finally, through a 'topographical' analysis of the co-polymer property space, we show how this large volume of data can be used to identify dominant monomers in specific regions of property space that may be amenable to a variety of applications, such as organic photovoltaics, light emitting diodes, and thermoelectrics.

From fine-grained properties to broad principles for gradual argumentation: A principled spectrum

The study of properties of gradual evaluation methods in argumentation has received increasing attention in recent years, with studies devoted to various classes of frameworks/ methods leading to conceptually similar but formally distinct properties in different contexts. In this paper we provide a novel systematic analysis for this research landscape by making three main contributions. First, we identify groups of conceptually related properties in the literature, which can be regarded as based on common patterns and, using these patterns, we evidence that many further novel properties can be considered. Then, we provide a simplifying and unifying perspective for these groups of properties by showing that they are all implied by novel parametric principles of (either strict or non-strict) balance and monotonicity. Finally, we show that (instances of) these principles (and thus the group, literature and novel properties that they imply) are satisfied by several quantitative argumentation formalisms in the literature, thus confirming the principles' general validity and utility to support a compact, yet comprehensive, analysis of properties of gradual argumentation.