Characteristic Attributes Research Articles

An Efficient Blockchain Assisted Reputation Aware Decentralized Federated Learning Framework

Because of the widespread presence and ease of access to the internet, edge devices are the perfect candidates for providing quality training on a variety of applications. However, their participation is restrained due to potential leakage of sensitive and private data. Federated learning targets to address these issues by bringing the model to the device and keeping the data in place. Still, it suffers from inherent security issues such as malicious participation and unfair contribution. The central server may become a bottleneck as well as induce biased aggregation and incentives. This article proposes a blockchain assisted federated learning framework, which fosters honest participation with reduced overheads, facilitating fair contribution-based weighted incentivization. A new consensus mechanism named PoIS (Proof of Interpretation and Selection) is proposed based on honest clients’ contributions. PoIS uses model interpretation techniques for evaluating and calculating individual contributions. The aggregation of feature attributions in PoIS, is able to detect the adversaries, and the label-wise aggregation of attributions across the participants helps to define the prominent contributors. Further, we devise a credit function based on the contribution, relevance as well as the past performance for calculating incentives. Extensive experiments have been carried out for the proposed architecture with different settings, models, and datasets, to verify our claim. It successfully restricts the attack to less than 5%, and selects the prominent (top-k) contributors. Theoretical analysis provides the guarantee for byzantine-robust aggregation, in a malicious setting.

Semi-Supervised Deep Learning-Based Multi-component Spectral Calibration Modeling for UV-vis and Near-Infrared Spectroscopy without Information Loss.

Spectral analysis is an important method for characterizing and identifying chemical species. However, quantitative spectral analysis of multiple chemical properties in the real world has always been a challenging problem due to the strong correlation, massive noise, and serious information overlapping of the spectral features. Here, we present a new semi-supervised spectral calibration method based on information lossless decoupling of spectral features named NICEM. To realize the separation and extraction of key latent features, the method uses the flow-based model non-linear independent component estimation (NICE) to learn the sample distribution. The spectral data information is transformed into independent latent variables obeying Gaussian distribution by the reversible structure of deep network without information loss, so as to find the essential properties and realize the feature nonlinear decomposition. Moreover, the association between the input latent feature variables and attributes is evaluated by the maximum mutual information coefficient to eliminate the adverse effects of irrelevant information in the latent variable space and mine key information. Since the latent variables are independent in each dimension, the NICEM method is easier to establish an accurate semi-supervised multi-component calibration model even for high overlapping and complex spectral data. The applicability of the proposed spectral modeling method is demonstrated by using three ultraviolet-visible and near-infrared spectral data sets with 15 physical and chemical properties including diesel fuels, corn, and multi-metal ions solution. Results show that the proposed NICEM method has the highest determination coefficient (R2) and significantly improves extrapolation compared with the seven state-of-the-art methods. The proposed method is intuitive because it obviates complex feature engineering and prior knowledge and is a promising spectral calibration tool for quantitative analysis in other spectroscopy applications.

MolSHAP: Interpreting Quantitative Structure-Activity Relationships Using Shapley Values of R-Groups.

Optimizing the activities and properties of lead compounds is an essential step in the drug discovery process. Despite recent advances in machine learning-aided drug discovery, most of the existing methods focus on making predictions for the desired objectives directly while ignoring the explanations for predictions. Although several techniques can provide interpretations for machine learning-based methods such as feature attribution, there are still gaps between these interpretations and the principles commonly adopted by medicinal chemists when designing and optimizing molecules. Here, we propose an interpretation framework, named MolSHAP, for quantitative structure-activity relationship analysis by estimating the contributions of R-groups. Instead of attributing the activities to individual input features, MolSHAP regards the R-group fragments as the basic units of interpretation, which is in accordance with the fragment-based modifications in molecule optimization. MolSHAP is a model-agnostic method that can interpret activity regression models with arbitrary input formats and model architectures. Based on the evaluations of numerous representative activity regression models on a specially designed R-group ranking task, MolSHAP achieved significantly better interpretation power compared with other methods. In addition, we developed a compound optimization algorithm based on MolSHAP and illustrated the reliability of the optimized compounds using an independent case study. These results demonstrated that MolSHAP can provide a useful tool for accurately interpreting the quantitative structure-activity relationships and rationally optimizing the compound activities in drug discovery.

Perceived Quality Analysis Using Importance Performance Analysis (IPA) Method in Teh Pucuk Harum

Perceived quality of a product is very important because it can be a factor in winning market competition. This study aims to determine what product attributes still need to be improved and which need to be maintained in the Teh Pucuk Harum packaged beverage product. The type of research used is mixed method. Data collection techniques using observation, surveys and interviews. As for data analysis using the Importance Performance Analysis (IPA) method. The results of the research show that the attributes of quality, design, and style still need to be improved and become the focus of the company. Meanwhile, price and feature attributes are considered less important by consumers, so they are not the focus of the company in its marketing strategy.

A user-centred evaluation of DisCERN: Discovering counterfactuals for code vulnerability detection and correction

Counterfactual explanations highlight actionable knowledge which helps to understand how a machine learning model outcome could be altered to a more favourable outcome. Understanding actionable corrections in source code analysis can be critical to proactively mitigate security attacks that are caused by known vulnerabilities. In this paper, we present the DisCERN explainer for discovering counterfactuals for code vulnerability correction. Given a vulnerable code segment, DisCERN finds counterfactual (i.e. non-vulnerable) code segments and recommends actionable corrections. DisCERN uses feature attribution knowledge to identify potentially vulnerable code statements. Subsequently, it applies a substitution-focused correction, suggesting suitable fixes by analysing the nearest-unlike neighbour. Overall, DisCERN aims to identify vulnerabilities and correct them while preserving both the code syntax and the original functionality of the code. A user study evaluated the utility of counterfactuals for vulnerability detection and correction compared to more commonly used feature attribution explainers. The study revealed that counterfactuals foster positive shifts in mental models, effectively guiding users towards making vulnerability corrections. Furthermore, counterfactuals significantly reduced the cognitive load when detecting and correcting vulnerabilities in complex code segments. Despite these benefits, the user study showed that feature attribution explanations are still more widely accepted than counterfactuals, possibly due to the greater familiarity with the former and the novelty of the latter. These findings encourage further research and development into counterfactual explanations, as they demonstrate the potential for acceptability over time among developers as a reliable resource for both coding and training.

Explainable artificial intelligence: A taxonomy and guidelines for its application to drug discovery

AbstractArtificial intelligence (AI) is having a growing impact in many areas related to drug discovery. However, it is still critical for their adoption by the medicinal chemistry community to achieve models that, in addition to achieving high performance in their predictions, can be trusty explained to the end users in terms of their knowledge and background. Therefore, the investigation and development of explainable artificial intelligence (XAI) methods have become a key topic to address this challenge. For this reason, a comprehensive literature review about explanation methodologies for AI based models, focused in the field of drug discovery, is provided. In particular, an intuitive overview about each family of XAI approaches, such as those based on feature attribution, graph topologies, or counterfactual reasoning, oriented to a wide audience without a strong background in the AI discipline is introduced. As the main contribution, we propose a new taxonomy of the current XAI methods, which take into account specific issues related with the typical representations and computational problems study in the design of molecules. Additionally, we also present the main visualization strategies designed for supporting XAI approaches in the chemical domain. We conclude with key ideas about each method category, thoroughly providing insightful analysis about the guidelines and potential benefits of their adoption in medical chemistry.This article is categorized under: Data Science > Artificial Intelligence/Machine Learning

Explaining compound activity predictions with a substructure-aware loss for graph neural networks.

Explainable machine learning is increasingly used in drug discovery to help rationalize compound property predictions. Feature attribution techniques are popular choices to identify which molecular substructures are responsible for a predicted property change. However, established molecular feature attribution methods have so far displayed low performance for popular deep learning algorithms such as graph neural networks (GNNs), especially when compared with simpler modeling alternatives such as random forests coupled with atom masking. To mitigate this problem, a modification of the regression objective for GNNs is proposed to specifically account for common core structures between pairs of molecules. The presented approach shows higher accuracy on a recently-proposed explainability benchmark. This methodology has the potential to assist with model explainability in drug discovery pipelines, particularly in lead optimization efforts where specific chemical series are investigated.

DC-SHAP Method for Consistent Explainability in Privacy-Preserving Distributed Machine Learning

Ensuring the transparency of machine learning models is vital for their ethical application in various industries. There has been a concurrent trend of distributed machine learning designed to limit access to training data for privacy concerns. Such models, trained over horizontally or vertically partitioned data, present a challenge for explainable AI because the explaining party may have a biased view of background data or a partial view of the feature space. As a result, explanations obtained from different participants of distributed machine learning might not be consistent with one another, undermining trust in the product. This paper presents an Explainable Data Collaboration Framework based on a model-agnostic additive feature attribution algorithm (KernelSHAP) and Data Collaboration method of privacy-preserving distributed machine learning. In particular, we present three algorithms for different scenarios of explainability in Data Collaboration and verify their consistency with experiments on open-access datasets. Our results demonstrated a significant (by at least a factor of 1.75) decrease in feature attribution discrepancies among the users of distributed machine learning. The proposed method improves consistency among explanations obtained from different participants, which can enhance trust in the product and enable ethical application in various industries.

Towards More Robust Interpretation via Local Gradient Alignment

Neural network interpretation methods, particularly feature attribution methods, are known to be fragile with respect to adversarial input perturbations. To address this, several methods for enhancing the local smoothness of the gradient while training have been proposed for attaining robust feature attributions. However, the lack of considering the normalization of the attributions, which is essential in their visualizations, has been an obstacle to understanding and improving the robustness of feature attribution methods. In this paper, we provide new insights by taking such normalization into account. First, we show that for every non-negative homogeneous neural network, a naive l2-robust criterion for gradients is not normalization invariant, which means that two functions with the same normalized gradient can have different values. Second, we formulate a normalization invariant cosine distance-based criterion and derive its upper bound, which gives insight for why simply minimizing the Hessian norm at the input, as has been done in previous work, is not sufficient for attaining robust feature attribution. Finally, we propose to combine both l2 and cosine distance-based criteria as regularization terms to leverage the advantages of both in aligning the local gradient. As a result, we experimentally show that models trained with our method produce much more robust interpretations on CIFAR-10 and ImageNet-100 without significantly hurting the accuracy, compared to the recent baselines. To the best of our knowledge, this is the first work to verify the robustness of interpretation on a larger-scale dataset beyond CIFAR-10, thanks to the computational efficiency of our method.

Rethinking Interpretation: Input-Agnostic Saliency Mapping of Deep Visual Classifiers

Saliency methods provide post-hoc model interpretation by attributing input features to the model outputs. Current methods mainly achieve this using a single input sample, thereby failing to answer input-independent inquiries about the model. We also show that input-specific saliency mapping is intrinsically susceptible to misleading feature attribution. Current attempts to use `general' input features for model interpretation assume access to a dataset containing those features, which biases the interpretation. Addressing the gap, we introduce a new perspective of input-agnostic saliency mapping that computationally estimates the high-level features attributed by the model to its outputs. These features are geometrically correlated, and are computed by accumulating model's gradient information with respect to an unrestricted data distribution. To compute these features, we nudge independent data points over the model loss surface towards the local minima associated by a human-understandable concept, e.g., class label for classifiers. With a systematic projection, scaling and refinement process, this information is transformed into an interpretable visualization without compromising its model-fidelity. The visualization serves as a stand-alone qualitative interpretation. With an extensive evaluation, we not only demonstrate successful visualizations for a variety of concepts for large-scale models, but also showcase an interesting utility of this new form of saliency mapping by identifying backdoor signatures in compromised classifiers.

Probably Approximate Shapley Fairness with Applications in Machine Learning

The Shapley value (SV) is adopted in various scenarios in machine learning (ML), including data valuation, agent valuation, and feature attribution, as it satisfies their fairness requirements. However, as exact SVs are infeasible to compute in practice, SV estimates are approximated instead. This approximation step raises an important question: do the SV estimates preserve the fairness guarantees of exact SVs? We observe that the fairness guarantees of exact SVs are too restrictive for SV estimates. Thus, we generalise Shapley fairness to probably approximate Shapley fairness and propose fidelity score, a metric to measure the variation of SV estimates, that determines how probable the fairness guarantees hold. Our last theoretical contribution is a novel greedy active estimation (GAE) algorithm that will maximise the lowest fidelity score and achieve a better fairness guarantee than the de facto Monte-Carlo estimation. We empirically verify GAE outperforms several existing methods in guaranteeing fairness while remaining competitive in estimation accuracy in various ML scenarios using real-world datasets.

Local Path Integration for Attribution

Path attribution methods are a popular tool to interpret a visual model's prediction on an input. They integrate model gradients for the input features over a path defined between the input and a reference, thereby satisfying certain desirable theoretical properties. However, their reliability hinges on the choice of the reference. Moreover, they do not exhibit weak dependence on the input, which leads to counter-intuitive feature attribution mapping. We show that path-based attribution can account for the weak dependence property by choosing the reference from the local distribution of the input. We devise a method to identify the local input distribution and propose a technique to stochastically integrate the model gradients over the paths defined by the references sampled from that distribution. Our local path integration (LPI) method is found to consistently outperform existing path attribution techniques when evaluated on deep visual models. Contributing to the ongoing search of reliable evaluation metrics for the interpretation methods, we also introduce DiffID metric that uses the relative difference between insertion and deletion games to alleviate the distribution shift problem faced by existing metrics. Our code is available at https://github.com/ypeiyu/LPI.

Bayesian-optimization-assisted discovery of stereoselective aluminum complexes for ring-opening polymerization of racemic lactide

Stereoselective ring-opening polymerization catalysts are used to produce degradable stereoregular poly(lactic acids) with thermal and mechanical properties that are superior to those of atactic polymers. However, the process of discovering highly stereoselective catalysts is still largely empirical. We aim to develop an integrated computational and experimental framework for efficient, predictive catalyst selection and optimization. As a proof of principle, we have developed a Bayesian optimization workflow on a subset of literature results for stereoselective lactide ring-opening polymerization, and using the algorithm, we identify multiple new Al complexes that catalyze either isoselective or heteroselective polymerization. In addition, feature attribution analysis uncovers mechanistically meaningful ligand descriptors, such as percent buried volume (%Vbur) and the highest occupied molecular orbital energy (EHOMO), that can access quantitative and predictive models for catalyst development.

GINT: A Generative Interpretability method via perturbation in the latent space

As the neural networks get deeper and deeper, model interpretation becomes more necessary and important, especially in high-risk fields such as medicine and finance. From the perspective of feature attribution, most existing attempts aim to identify relevant features contributing the most to the prediction. Among them, perturbation-based methods mainly explore the corresponding model’s output by randomly perturbing the given features. When the data is high-dimensional and sparse, perturbing the feature space may be inefficient and meaningless since it ignores the feature correlations in the data distribution. In this paper, we introduce a novel Generative INTerpretability method, named GINT, which generates perturbations in the latent space. We propose a unified framework for perturbation-based methods, which describes the characteristics of a suitable perturbation for interpretation. Under the framework, we adopt a generative model to generate perturbation instead of randomly perturbing. Subsequently, we sample perturbations from the generative model for a given instance and its prediction and calculate the feature importance by analyzing those perturbations. We conduct extensive experiments to validate the effectiveness and efficiency of the proposed method.

Post-blast capacity evaluation of concrete-filled steel tubular (CFST) column based on machine learning technique

In the present study, the post-blast axial load-bearing capacity of the CFST column under the close-in explosion is investigated. Dynamic response and residual axial load capacity (RALBC) of 1599 circular CFST columns are numerically studied using a validated finite element model. The effects of column dimensions, material properties, blast loading, and axial load ratio are discussed. Based on the huge database, prediction models of RALBC are developed using three commonly used machine learning techniques including Extreme Gradient Boosting (known as XGBoost), Artificial Neutral Network (ANN), and Support Vector Regression (SVR). Comparison results show that the prediction model generated by the XGBoost performs the best among the three methods followed by the ANN and SVR methods. The RALBC of the CFST columns can be estimated rapidly and accurately with the prediction model developed in the present study. Finally, the feature importance of the input variables is analyzed using the additive feature attribution method SHAP (Shapley Additive ExPlanation). The results show that geometric parameters are the main factors impacting the RALBC of the CFST columns under close-in explosion.

PERCY: A post-hoc explanation-based score for logic rule dissemination consistency assessment in sentiment classification

Disseminating and incorporating logic rules into deep neural networks has been extensively explored for sentiment classification in recent years. In particular, most methods and algorithms proposed for this purpose rely on a specific component that aims to capture and model logic rules, followed by a sequence model to process the input sequence. While the authors of these methods claim that they effectively capture syntactic structures that affect sentiment classification, they only show improvement in accuracy to support their claims without further analysis. Focusing on various syntactic structures, particularly contrastive discourse relations such as the A-but-B structure, we introduce the PERCY score, a novel Post-hoc Explanation-based Rule ConsistencY Score to analyze and study the ability of several of these methods to identify these structures in a given sentence, and to make their classification decisions based on the appropriate conjunct. Specifically, we explore the use of model-agnostic post-hoc explanation frameworks to explain the predictions of any classifier in an interpretable and faithful manner. These model explainability frameworks provide feature attribution scores to estimate each word’s impact on the final classification decision. Then, they are combined to check whether the model has based its decision on the right conjunct. Our experiments show that (a) accuracy – or any other performance metric – can be misleading in assessing the ability of logic rule dissemination methods to base their decisions on the right conjunct, (b) not all analyzed methods effectively capture syntactic structures, (c) often, the underlying sequence model is what captures the structure, and (d) for the best method, less than 25% of the test examples are classified based on the appropriate conjunct, indicating that a lot of research needs to be done on this topic. Finally, we experimentally demonstrate that the PERCY scores calculated are robust and stable w.r.t. the feature-attribution frameworks used.

Radiobioinformatics: A novel bridge between basic research and clinical practice for clinical decision support in diffuse liver diseases

AbstractThe liver is a multifaceted organ that is responsible for many critical functions encompassing amino acid, carbohydrate, and lipid metabolism, all of which make a healthy liver essential for the human body. Contemporary imaging methodologies have remarkable diagnostic accuracy in discerning focal liver lesions; however, a comprehensive understanding of diffuse liver diseases is a requisite for radiologists to accurately diagnose or predict the progression of such lesions within clinical contexts. Nonetheless, the conventional attributes of radiological features, including morphology, size, margin, density, signal intensity, and echoes, limit their clinical utility. Radiomics is a widely used approach that is characterized by the extraction of copious image features from radiographic depictions, which gives it considerable potential in addressing this limitation. It is worth noting that functional or molecular alterations occur significantly prior to the morphological shifts discernible by imaging modalities. Consequently, the explication of potential mechanisms by multiomics analyses (encompassing genomics, epigenomics, transcriptomics, proteomics, and metabolomics) is essential for investigating putative signal pathway regulations from a radiological viewpoint. In this review, we elaborate on the principal pathological categorizations of diffuse liver diseases, the evaluation of multiomics approaches pertaining to diffuse liver diseases, and the prospective value of predictive models. Accordingly, the overarching objective of this review is to scrutinize the interrelations between radiological features and bioinformatics as well as to consider the development of prediction models predicated on radiobioinformatics as integral components of clinical decision support systems for diffuse liver diseases.

Regularized Pairwise Relationship based Analytics for Structured Data

In line with the increasing machine learning model inference accuracy, deep learning (DL) models have been increasingly applied to structured data for a wide spectrum of real-world applications, including product recommendations, online advertisement, healthcare analytics and risk analysis. However, unlike unstructured data, structured data is high-dimensional and sparse and therefore engenders a large number of parameters in DL, making DL models more prone to overfitting. To alleviate the overfitting problem, various regularization methods have been designed to constrain the model parameters as a means to control the model complexity. Unfortunately, these methods are often restricted to regularizing the parameter values directly without considering the intrinsic correlations and dependencies between attribute fields of structured data which is however key to effective structured data modeling. In this paper, we re-examine DL for structured data from a new perspective of attribute interactions. In particular, we seek to explicitly model and regularize the pairwise relationships between attribute fields of structured data, in a field-adaptive manner, via a proposed attentive and interpretable framework called ATT-Reg. Specifically, in this framework, a set of attentive weight matrices are introduced to each attribute field for modeling obviously different relationships with its neighboring attribute fields. Further, we derive from the Bayesian viewpoint a novel Attentive Regularization method for imposing adaptive regularization strengths on different pairs of attribute fields, based on the informativeness of their relationship, which is calculated using both data-driven information and functional dependency (FD) knowledge. Such adaptive regularization facilitates each attribute field to learn discriminative and diversified representations for more effective predictive analytics. We also develop a feature attribution method for supporting more interpretable predictions We validate the effectiveness of our ATT-Reg on six real-world datasets. Extensive experimental results show that ATT-Reg achieves significant improvement over state-of-the-art graph models, attentive models as well as regularization methods and supports an excellent degree of interpretation.

Algorithms to estimate Shapley value feature attributions

Feature attributions based on the Shapley value are popular for explaining machine learning models. However, their estimation is complex from both theoretical and computational standpoints. We disentangle this complexity into two main factors: the approach to removing feature information and the tractable estimation strategy. These two factors provide a natural lens through which we can better understand and compare 24 distinct algorithms. Based on the various feature-removal approaches, we describe the multiple types of Shapley value feature attributions and the methods to calculate each one. Then, based on the tractable estimation strategies, we characterize two distinct families of approaches: model-agnostic and model-specific approximations. For the model-agnostic approximations, we benchmark a wide class of estimation approaches and tie them to alternative yet equivalent characterizations of the Shapley value. For the model-specific approximations, we clarify the assumptions crucial to each method’s tractability for linear, tree and deep models. Finally, we identify gaps in the literature and promising future research directions. There are numerous algorithms for generating Shapley value explanations. The authors provide a comprehensive survey of Shapley value feature attribution algorithms by disentangling and clarifying the fundamental challenges underlying their computation.

Extracting principal building variables from automatically collected urban scale façade images for energy conservation through deep transfer learning

Buildings account for 40% of the energy consumption and 13% of the greenhouse gas (GHG) emissions in the U.S. To improve building energy efficiency, cities around the U.S. issued energy policies, such as the energy performance disclosure requirement in New York City, to encourage building owners to make informed retrofitting decisions. However, complying with these policies is expensive and time-consuming for government agencies and building owners, especially for old buildings where detailed building information is not readily available. In this work, we propose an automatic, non-intrusive, and scalable framework to capture energy-essential building variables through reasoning building façade images - FaçadeReasoner. Specifically, we first build a comprehensive building information dataset and identify the most impactful (i.e., principal) variables in relation to building energy performance and GHG emissions using the state-of-the-art feature attribution model. Next, we propose a method to automatically collect an urban scale building image dataset with more than 10,000 façade images and extract principal-building-variables from these images using deep transfer learning. Results show that FacadeReasoner has the capability to predict principal building variables, namely “building type” (accuracy 0.77), “year built” (accuracy 0.62), “building height” (R2 0.80), and rough estimates of “building area” (R2 0.46) from façade images. This study is unique as it marks the first attempt to enable an automated end-to-end framework for urban scale principal-building-variables extraction, providing an efficient and economical alternative for large building portfolio owners and managers (e.g., municipalities) to comprehend urban scale energy-related building information for informed decision-making, directly contributing to Net-Zero 2050.