Hard Cases Research Articles

CED: A case-level explainable paramedical diagnosis via AdaGBDT

Objective:The rapid growth of medical data has greatly promoted the wide exploitation of machine learning for paramedical diagnosis. Inversely proportional to their performance, most machine learning models generally suffer from the lack of explainability, especially the local explainability of the model, that is, the case-specific explainability. Materials and Methods:In this paper, we proposed a GBDT (Gradient Boosting Decision Tree)-based explainable model for case-specific paramedical diagnostics, and mainly make the following contributions: (1) an adaptive gradient boosting decision tree (AdaGBDT) model is proposed to boost the path-mining for decision effectively; (2) to learn a case-specific feature importance embedding for a specific patient, the bi-side mutual information is applied to characterize the backtracking on the decision path; (3) through the collaborative decision-making by globally explainable AdaGBDT with case-based reasoning (CBR) in the case-specific metric space, some hard cases can be identified by the means of visualized interpretation. The performance of our model is evaluated on the Wisconsin diagnostic breast cancer dataset and the UCI heart disease dataset. Results:Experiments conducted on two datasets show that our AdaGBDT achieves the best performance, with the F1-value of 0.9647 and 0.8405 respectively. Moreover, a series of experimental analyses and case studies further illustrate the excellent performance of feature importance embedding. Conclusion:The proposed case-specific explainable paramedical diagnosis via AdaGBDT has excellent predictive performance, with both promising case-level and consistent global explainability.

The Aboriginal Shield from the Collection of the British Museum: A Case Study from the Perspective of Recent Developments

This article analyses the case of the dispute over the return of the Aboriginal shield from the collection of the British Museum, which up to 2018 was believed to have come from the First Contact with the Aboriginal people made by the crew of captain James Cook. The arguments exchanged between the parties are studied from the perspective of the theory of hard cases. The claim for the return expressed by Rodney Kelly is supported mostly by the arguments based on justice, and both personal and cultural affiliation. The British Museum responded to the claim with results of research questioning the provenance of the shield. The relevance of the research is evaluated in the context of the symbolic significance of the case. The author believes that application of the theory of hard cases allows to broaden the debate beyond the scope of positive law.

Eco-relational Pluralism: Political Liberalism’s Challenge to the Economic Growth Imperative

ABSTRACT Rawls theorizes principles of justice as defining a ‘pact of reconciliation’ between diverse conceptions of the good. What does fulfillment of this pact entail when reasonable pluralism is recognized as having an environmental dimension? Fair acknowledgment of the plurality of citizens’ relationships with the natural world challenges the neutrality of aims conventionally used to justify ecocide, including the promotion of economic growth and development. This paper explores how ecocide constitutes a violation of equal basic liberties and state neutrality as per Rawlsian political liberalism. It further considers hard cases in which ecocide is enacted for meeting citizens’ basic needs.

Meta’s Oversight Board and Beyond – Corporations as Interpreters and Adjudicators of International Human Rights

Abstract Social media platform corporations such as Meta (Facebook), Twitter, etc. find themselves in a position of having to interpret international human rights norms, in particular Article 19 of the International Covenant on Civil and Political Rights (ICCPR) (freedom of expression). Millions or even billions of content moderation decisions need to be taken on the platforms each day that affect users’ human rights interests. Since content moderation is integral to the technical and commercial set-up of these platforms, corporate decision-making vis-à-vis human rights and thus corporate interpretation of international human rights norms is inevitable. Yet, corporations are flawed interpreters. Whereas they act, like a court or tribunal, as triadic decision-makers, they, unlike a court or tribunal, do not share the neutrality, impartiality and independence of the latter. In particular, they are responsible to their shareholders and they pursue commercial interests when moderating content. This article grapples with the theoretical and doctrinal implications of flawed but inevitable corporate human rights interpretation. Taking the early practice of the Oversight Board, a body established by Meta, Inc. (Facebook), in order to tackle the “hard cases” of content moderation as a case study, the pitfalls and challenges of corporate human rights interpretation become apparent. In the end, I submit a few suggestions in order to remedy what seems to be here to stay with us at least for as long as an important part of public discourse is channelled through social media platforms.

TwoViewDensityNet: Two-View Mammographic Breast Density Classification Based on Deep Convolutional Neural Network

Dense breast tissue is a significant factor that increases the risk of breast cancer. Current mammographic density classification approaches are unable to provide enough classification accuracy. However, it remains a difficult problem to classify breast density. This paper proposes TwoViewDensityNet, an end-to-end deep learning-based method for mammographic breast density classification. The craniocaudal (CC) and mediolateral oblique (MLO) views of screening mammography provide two different views of each breast. As the two views are complementary, and dual-view-based methods have proven efficient, we use two views for breast classification. The loss function plays a key role in training a deep model; we employ the focal loss function because it focuses on learning hard cases. The method was thoroughly evaluated on two public datasets using 5-fold cross-validation, and it achieved an overall performance (F-score of 98.63%, AUC of 99.51%, accuracy of 95.83%) on DDSM and (F-score of 97.14%, AUC of 97.44%, accuracy of 96%) on the INbreast. The comparison shows that the TwoViewDensityNet outperforms the state-of-the-art methods for classifying breast density into BI-RADS class. It aids healthcare providers in providing patients with more accurate information and will help improve the diagnostic accuracy and reliability of mammographic breast density evaluation in clinical care.

The Age of Johnson, ed. Jack Lynch and J. T. Scanlan

<i>The Age of Johnson</i>, ed. Jack Lynch and J. T. Scanlan

Is the European Charter of Local Self-Government an Effective Instrument for the Protection of Local Autonomy in Poland?

Purpose: The objective of this research is to examine whether the European Charter of Local Self-Government is, in practice, an effective instrument for the protection of the autonomy of local government in Poland, as well as well as to define the contribution of the Polish case law and administrative practice to the understanding of the principles of the Charter. The importance of the subject is strengthened by the political context. For many years, Poland was considered a model country safeguarding extensive local autonomy. Recently, however, recentralization trends have emerged in government policies. Design/methodology/approach: The research is based on the qualitative and quantitative empirical research of the case law and administrative practice with elements of doctrinal analysis.Findings: The Charter is present in the case law of the Constitutional Tribunal (22 judgments), administrative courts (166), and public administration bodies supervising local governments (49). The number is high compared to other CoE countries. The administrative courts seem to be more eager to adjudicate in favour of local governments in the cases in which the Charter is referred to.Academic contribution to the field: A specific Polish input to the application of the Charter is the frequent use of the principle of proportionality in the supervision of local governments, which serves the courts as a perfect tool for resolving ‘hard cases’ between the local government and central administration bodies. In several important judgments, Article 11 concerning the judicial protection of local governments was invoked to effectively strike down the law limiting the local governments’ access to court in specific cases.Practical implications: The research may be useful for local government representatives in formulating their litigation strategies, especially in legal disputes with central authorities.Originality/value: This research is the first all-embracing empirical research of the application of the Charter in a country’s legal practice. It shows original ways of the practical use of the Charter not described in the literature so far.

Algebraic hyperbolicity of very general surfaces

Recently, Haase and Ilten initiated the study of classifying algebraically hyperbolic surfaces in toric threefolds. We complete this classification for $${\mathbb{P}^1} \times {\mathbb{P}^1} \times {\mathbb{P}^1},\,\,{\mathbb{P}^2} \times {\mathbb{P}^1},\,\,{\mathbb{F}_e} \times {\mathbb{P}^1}$$ and the blowup of ℙ3 at a point, augmenting our earlier work on ℙ3. Most importantly, we treat the boundary cases which are the hardest cases from the point of view of positivity. In the process, we codify several different techniques for proving algebraic hyperbolicity, allowing us to prove similar results for hypersurfaces in any threefold admitting a group action with dense orbit.

Some Assembly Required: Explaining Variations in Legislative Oversight over the Armed Forces

AbstractLegislatures vary widely in how they affect democratic civil–military relations. In some countries, legislative oversight plays a critical role in guiding their defense establishment. In others, legislators are largely ignorant and happily so. In this article, we explain the sources of these variations in fifteen democratic states. After discussing the importance of the legislature's role in democratic civil–military relations, we clarify what we mean by oversight. We argue that variations in oversight are explained by the number and scope of legislative committees charged with military oversight and party politics within those committees. After reviewing alternative explanations, we present oversight patterns in fifteen democratic countries across the world. We then briefly examine Germany's Bundestag and Japan's Diet, as the comparison of these cases challenges most existing explanations of legislative oversight and serve as hard cases for our argument. We conclude with implications of legislative oversight for broader debates about civilian control of the military.

Subsidiarity does not win cases: A mixed methods study of the relationship between margin of appreciation language and deference at the European Court of Human Rights

AbstractIn August 2021, Protocol 15 inserted the doctrine of the margin of appreciation into the preamble of the European Convention of Human Rights, presumably cementing what President Spano has referred to as the ‘Age of Subsidiarity’, in which the European Court of Human Rights applies the margin of appreciation more often and increases deference to state parties. This insertion was done on the behest of the High Contracting Parties as part of the Interlaken reform process, and there is already a strong narrative in certain member states and parts of the scholarly literature that this focus has prompted the Court to increase the usage of the margin of appreciation and therefore the deference to states, judging more frequently in their favour. This article hypothesizes, however, that the increased usage of the margin of appreciation language which has been taken as proof for this narrative, might not, in fact, indicate higher levels of deference. Rather, the language of the margin of appreciation could be the result of usage by other actors or a marker of complexity for so-called ‘hard cases’. To investigate this relationship, the article applies a mixed legal-doctrinal and quantitative methodology to analyse who in the case law invokes the doctrine, what their purpose is for doing so, and what adjudicative consequences follow. It finds that usage of the margin-language topped well before the Interlaken process began, that governments are not the most frequent invokers and that, statistically speaking, states are no more likely to win margin-cases than other cases.

Property guardians: reigniting the lease/licence battle?

This article analyses recent English decisions reviving the need to consider the lease/licence dichotomy and conclusiveness of the parties’ agreement in the new context of property guardianship as an alternative to private renting. It argues that context has proved instructive in interpreting the parties’ agreement elsewhere in the case law and offers a way forward in the hard cases amid the ongoing search for doctrinal clarity and justification. A compound subjective–objective approach appreciates the underlying purpose of the parties’ relationship and justifies why no intention to grant the right of exclusive possession can be present, thereby precluding a tenancy. The article briefly considers reforms to rental accommodation previously suggested by the Law Commission and, in light of the continued need to prove the status of lessee, argues that they should be revisited in order to protect those living in temporary accommodation.

Compound Figure Separation of Biomedical Images: Mining Large Datasets for Self-supervised Learning

With the rapid development of self-supervised learning (e.g., contrastive learning), the importance of having large-scale images (even without annotations) for training a more generalizable AI model has been widely recognized in medical image analysis. However, collecting large-scale task-specific unannotated data at scale can be challenging for individual labs. Existing online resources, such as digital books, publications, and search engines, provide a new resource for obtaining large-scale images. However, published images in healthcare (e.g., radiology and pathology) consist of a considerable amount of compound figures with subplots. In order to extract and separate compound figures into usable individual images for downstream learning, we propose a simple compound figure separation (SimCFS) framework without using the traditionally required detection bounding box annotations, with a new loss function and a hard case simulation. Our technical contribution is four-fold: (1) we introduce a simulation-based training framework that minimizes the need for resource extensive bounding box annotations; (2) we propose a new side loss that is optimized for compound figure separation; (3) we propose an intra-class image augmentation method to simulate hard cases; and (4) to the best of our knowledge, this is the first study that evaluates the efficacy of leveraging self-supervised learning with compound image separation. From the results, the proposed SimCFS achieved state-of-the-art performance on the ImageCLEF 2016 Compound Figure Separation Database. The pretrained self-supervised learning model using large-scale mined figures improved the accuracy of downstream image classification tasks with a contrastive learning algorithm. The source code of SimCFS is made publicly available at &lt;a href='https://github.com/hrlblab/ImageSeperation'&gt;https://github.com/hrlblab/ImageSeperation&lt;/a&gt;

Good Clients and Hard Cases: The Role of Typologies at the Welfare Front Line

Frontline staff play a critical role in welfare-to-work delivery, making decisions about who gets selected for which supports and sanctions. Research on these street-level bureaucrats shows them rarely implementing policy exactly as written. Instead, they bring differing valuation frames and identity assessments to this work, resulting in different implementations of policy. Central to this selectivity is the way they type and classify different clients’ characteristics. Using case data from four Australian employment services sites collected over 18 months of fieldwork, we explore how frontline staff members sort clients according to two important dimensions of their perceived conduct: job readiness as workers and trustworthiness as people. We examine the staff’s rationale and the implications of these categorization practices for how clients are treated.

Fast Parallel Evaluation of Exact Geometric Predicates on GPUs

This paper accelerates the exact evaluation of large numbers of 3D geometric predicates with an algorithm whose work is partitioned between the CPU and the GPU on a high-performance computer to exploit the relative strengths of each. The test algorithm computes all the red–blue intersections between a set of red 3D triangles and another set of blue 3D triangles. A sequence of filters is employed that progressively eliminates more and more red–blue pairs that do not intersect, finally leaving only the actual intersections. Initially, a uniform grid is constructed on the GPU to identify pairs of nearby triangles. Then, these pairs are tested for intersection with single-precision interval arithmetic on the GPU. The ambiguous cases are next filtered with double-precision interval arithmetic on the multi-core CPU, and finally the hard cases are re-evaluated in parallel on the CPU using arbitrary-precision rational numbers. The parallel speedup for the whole algorithm was up to 414 times. It took only 1.17 s to find the 18M intersections between two datasets containing a total of 14M triangles. The intersection computation was sped up by up to 1936 times. The techniques that gave this excellent performance should be useful for parallelizing other geometric algorithms in fields such as CAD, GIS, and 3D modeling.

Graph Coloring via Clique Search with Symmetry Breaking

It is known that the problem of proper coloring of the nodes of a given graph can be reduced to finding cliques in a suitably constructed auxiliary graph. In this work, we explore the possibility of reducing the search space by exploiting the symmetries present in the auxiliary graph. The proposed method can also be used for efficient exact coloring of hyper graphs. We also precondition the auxiliary graph in order to further reduce the search space. We carry out numerical experiments to assess the practicality of these proposals. We solve some hard cases and prove a new lower limit of seven for the mycielski7 graph with the aid of the proposed technique.

Difficult Capacity Cases-The Experience of Liaison Psychiatrists. An Interview Study Across Three Jurisdictions.

BackgroundAssessment of capacity for treatment and discharge decisions is common in the general hospital. Liaison psychiatrists are often asked to support the treating medical or surgical team in difficult capacity assessments. However, empirical research on identification and resolution of difficult capacity cases is limited. Some studies have identified certain patient, decisional, and interpersonal factors which cause difficulty, but no study has explored how these issues are resolved in practice. Our study therefore aimed to describe how experienced liaison psychiatrists identify and resolve difficult capacity cases in a general hospital setting.MethodsWe carried out semi-structured interviews with 26 liaison psychiatrists from England, Scotland, and New Zealand, on their most difficult capacity cases. Thematic analysis was used to examine types of difficulty and how these were resolved in practice. Summaries were prepared and example quotes extracted to illustrate phenomena described.ResultsWe identified four types of difficulty in capacity assessment, spanning both clinical and ethical domains: 1) Difficulty determining whether the decision is the patient's own or driven by illness, 2) Difficulty in applying ethical principles, 3) Difficulty in avoiding personal bias, and 4) Procedural difficulties. The liaison psychiatrists presented as self-reflective and aware of challenges and pitfalls in hard cases. We summarized their creative strategies to resolve difficulty in assessment.ConclusionPractitioners approaching difficult capacity cases require both clinical skills, e.g., to uncover subtle illness impairing decision-making and to consider interpersonal dynamics, and ethical skills, e.g., to negotiate the role of values and risks in capacity assessment. Education and training programmes should incorporate both aspects and could include the resolution strategies identified in our study. Practitioners, supported by health and social care systems, should work to develop self-aware and reflective capacity assessment practice.

Visibility of points: Mining occlusion cues for monocular 3D object detection

Monocular 3D object detection aims at achieving prediction from two-dimensional image plane to three-dimensional physical world. It is an inevitable problem that occlusion phenomena limit the performance in practice. To solve the challenging problem that directly represents the spatial information of occlusion relation, we propose the visibility states of points to describe the spatial distance relationships of occlusion pairs and the implied orientation information. The visibility state introduction can better represent the level and direction of occlusion information and enhance the network’s understanding of occlusion information. Furthermore, we redesign an end-to-end detector to encode features of visibility states to integrate occlusion ordering cues of the whole image to assist object localization in world space. Experiments on the KITTI3D dataset indicate that our method succeeds in establishing visibility states as occlusion cues and promoting the performance of the original detector. Our method is effective, and the performance is comparable with state-of-the-art approaches, especially outstanding in Moderate and Hard cases. Specifically, our method improves the accuracy of 3D moderate case detection to 42.75% and hard case to 37.03% in the KITTI3D dataset.

Sim2Real Object-Centric Keypoint Detection and Description

Keypoint detection and description play a central role in computer vision. Most existing methods are in the form of scene-level prediction, without returning the object classes of different keypoints. In this paper, we propose the object-centric formulation, which, beyond the conventional setting, requires further identifying which object each interest point belongs to. With such fine-grained information, our framework enables more downstream potentials, such as object-level matching and pose estimation in a clustered environment. To get around the difficulty of label collection in the real world, we develop a sim2real contrastive learning mechanism that can generalize the model trained in simulation to real-world applications. The novelties of our training method are three-fold: (i) we integrate the uncertainty into the learning framework to improve feature description of hard cases, e.g., less-textured or symmetric patches; (ii) we decouple the object descriptor into two independent branches, intra-object salience and inter-object distinctness, resulting in a better pixel-wise description; (iii) we enforce cross-view semantic consistency for enhanced robustness in representation learning. Comprehensive experiments on image matching and 6D pose estimation verify the encouraging generalization ability of our method. Particularly for 6D pose estimation, our method significantly outperforms typical unsupervised/sim2real methods, achieving a closer gap with the fully supervised counterpart.

Constraint Satisfaction Problems with Global Modular Constraints: Algorithms and Hardness via Polynomial Representations

We study the complexity of Boolean constraint satisfaction problems (CSPs) when the assignment must have Hamming weight in some congruence class modulo $M$, for various choices of the modulus $M$. Due to the known classification of tractable Boolean CSPs, this mainly reduces to the study of three cases: 2-SAT, HORN-SAT, and LIN-2 (linear equations mod 2). We classify the moduli $M$ for which these respective problems are polynomial time solvable, and when they are not (assuming the exponential time hypothesis). Our study reveals that this modular constraint lends a surprising richness to these classic, well-studied problems, with interesting broader connections to complexity theory and coding theory. The HORN-SAT case is connected to the covering complexity of polynomials representing the NAND function mod $M$. The LIN-2 case is tied to the sparsity of polynomials representing the OR function mod $M$, which in turn has connections to modular weight distribution properties of linear codes and locally decodable codes. In both cases, the analysis of our algorithm as well as the hardness reduction rely on these polynomial representations, highlighting an interesting algebraic common ground between hard cases for our algorithms and the gadgets which show hardness. These new complexity measures of polynomial representations merit further study. The inspiration for our study comes from a recent work by Nägele, Sudakov, and Zenklusen on submodular minimization with a global congruence constraint. Our algorithm for HORN-SAT has strong similarities to their algorithm, and in particular identical kinds of set systems arise in both cases. Our connection to polynomial representations leads to a simpler analysis of such set systems and also sheds light on (but does not resolve) the complexity of submodular minimization with a congruency requirement modulo a composite $M$.

Algorithms and Complexities of Matching Variants in Covariate Balancing

In an observational study there are two disjointed groups of samples, one of treatment samples and the other of control samples. Each of the samples is characterized by several observed covariates. Covariate balancing problems arise when estimating causal effects using observational data. It is desirable to replicate a randomized experiment by obtaining treatment and control groups with similar covariate distributions. Even though covariate balancing problems have been studied extensively, the complexity status of many variants has not been established. Some of our results demonstrate that 2-covariate balancing problems are polynomial time solvable, whereas almost all problems are hard for three or more covariates. These results have practical implications, such as justifying the use of implicit enumeration techniques or heuristics for the hard cases. A new approach suggested by our results is to use the 2-covariate polynomial cases and relax the problem by aggregating covariates into two sets to be solved efficiently. (accepted paper title: Algorithms and complexities of matching variants in covariate balancing)