Intuitive Strategy Research Articles

Answering, Fast and Slow: Strategy enhancement of visual understanding guided by causality

In his classic book Thinking, Fast and Slow (Daniel, 2017), Kahneman points out that human thinking can be categorized into two main modes of thinking: a system that displays intuition and emotion (i.e., System 1), and a system that is more planned and relies more on logic, defined as System 2. This idea explains both rational and irrational motivations. In this paper, we revisit visual comprehension tasks based on this idea. At the theoretical level, we focus on the relationship between intuitive thinking, prior knowledge, and environmental information, and build a causal graph between the three. Further, inspired by the constructed causal graph, an intuitive optimization strategy with clear interpretability is proposed. In the validation session, we provide conclusions consistent with the theoretical analyses through extensive experiments on public datasets based on a visual quizzing task. Excitingly, our scheme demonstrates strong competitiveness in terms of generalizability without adding new technologies.

Minimizing Oxidative Stress in the Lens: Alternative Measures for Elevating Glutathione in the Lens to Protect against Cataract.

Oxidative stress plays a major role in the formation of the cataract that is the result of advancing age, diabetes or which follows vitrectomy surgery. Glutathione (GSH) is the principal antioxidant in the lens, and so supplementation with GSH would seem like an intuitive strategy to counteract oxidative stress there. However, the delivery of glutathione to the lens is fraught with difficulties, including the limited bioavailability of GSH caused by its rapid degradation, anatomical barriers of the anterior eye that result in insufficient delivery of GSH to the lens, and intracellular barriers within the lens that limit delivery of GSH to its different regions. Hence, more attention should be focused on alternative methods by which to enhance GSH levels in the lens. In this review, we focus on the following three strategies, which utilize the natural molecular machinery of the lens to enhance GSH and/or antioxidant potential in its different regions: the NRF2 pathway, which regulates the transcription of genes involved in GSH homeostasis; the use of lipid permeable cysteine-based analogues to increase the availability of cysteine for GSH synthesis; and the upregulation of the lens's internal microcirculation system, which is a circulating current of Na+ ions that drives water transport in the lens and with it the potential delivery of cysteine or GSH. The first two strategies have the potential to restore GSH levels in the epithelium and cortex, while the ability to harness the lens's internal microcirculation system offers the exciting potential to deliver and elevate antioxidant levels in its nucleus. This is an important distinction, as the damage phenotypes for age-related (nuclear) and diabetic (cortical) cataract indicate that antioxidant delivery must be targeted to different regions of the lens in order to alleviate oxidative stress. Given our increasing aging and diabetic populations it has become increasingly important to consider how the natural machinery of the lens can be utilized to restore GSH levels in its different regions and to afford protection from cataract.

MNN: Mixed nearest-neighbors for self-supervised learning

In contrastive self-supervised learning, positive samples are typically drawn from the same image but in different augmented views, resulting in a limited source of positive samples. An effective way to alleviate this problem is to incorporate the relationship between samples, which involves including the top-K nearest neighbors of positive samples. However, the issue of false neighbors (i.e., neighbors that do not belong to the same category as the positive sample) is a significant yet often overlooked challenge. In this paper, we present a simple self-supervised learning framework called Mixed Nearest-Neighbors (MNN) for Self-Supervised Learning. The primary objective of the MNN is to enhance the robustness and accuracy of the model by strategically incorporating synthesized neighboring samples. Specifically, our proposed method utilizes image mixture to mitigate the effects of noise introduced by false neighbors, while adopting an intuitive weighting strategy that effectively integrates these synthesized neighbors into the model. The linear evaluation of MNN on CIFAR-10, CIFAR,100, STL-10, and Tiny ImageNet shows an improvement of 0.41% to 1.96% over the previous state-of-the-art (SOTA) methods and 1.82% to 8.8% over the Mean Shift (MSF) method. Furthermore, the proposed components can be seamlessly integrated into other self-supervised learning algorithms to enhance their performance. Code is available at https://github.com/pc-cp/MNN.

User Evaluation of a Shared Robot Control System Combining BCI and Eye Tracking in a Portable Augmented Reality User Interface.

This study evaluates an innovative control approach to assistive robotics by integrating brain-computer interface (BCI) technology and eye tracking into a shared control system for a mobile augmented reality user interface. Aimed at enhancing the autonomy of individuals with physical disabilities, particularly those with impaired motor function due to conditions such as stroke, the system utilizes BCI to interpret user intentions from electroencephalography signals and eye tracking to identify the object of focus, thus refining control commands. This integration seeks to create a more intuitive and responsive assistive robot control strategy. The real-world usability was evaluated, demonstrating significant potential to improve autonomy for individuals with severe motor impairments. The control system was compared with an eye-tracking-based alternative to identify areas needing improvement. Although BCI achieved an acceptable success rate of 0.83 in the final phase, eye tracking was more effective with a perfect success rate and consistently lower completion times (p<0.001). The user experience responses favored eye tracking in 11 out of 26 questions, with no significant differences in the remaining questions, and subjective fatigue was higher with BCI use (p=0.04). While BCI performance lagged behind eye tracking, the user evaluation supports the validity of our control strategy, showing that it could be deployed in real-world conditions and suggesting a pathway for further advancements.

START: A Versatile Platform for Bacterial Ligand Sensing with Programmable Performances.

Recognition of signaling molecules for coordinated regulation of target genes is a fundamental process for biological systems. Cells often rely on transcription factors to accomplish these intricate tasks, yet the subtle conformational changes of protein structures, coupled with the complexity of intertwined protein interaction networks, pose challenges for repurposing these for bioengineering applications. This study introduces a novel platform for ligand-responsive gene regulation, termed START (Synthetic Trans-Acting Riboswitch with Triggering RNA). Inspired by the bacterial ligand sensing system, riboswitch, and the synthetic gene regulator, toehold switch, the START platform enables the implementation of synthetic biosensors for various ligands. Rational sequence design with targeted domain optimization yields high-performance STARTs with a dynamic range up to 67.29-fold and a tunable ligand sensitivity, providing a simple and intuitive strategy for sensor engineering. The START platform also exhibits modularity and composability to allow flexible genetic circuit construction, enabling seamless implementation of OR, AND, and NOT Boolean logic gates for multiple ligand inputs. The START design principle is capable of broadening the suite of synthetic biosensors for diverse chemical and protein ligands, providing a novel riboregulator chassis for synthetic biology and bioengineering applications.

Learning cardiac embryology on YouTube-What videos are there to view?

In modern medical curricula, embryology is typically taught through lectures, with a few institutions providing tutorials. The use of 3-D videos or animations enables students to study these embryological structures and how they change with time. The aim of this study was to assess the quality of cardiac embryology videos available on YouTube. A systematic literature review regarding the use of YouTube in teaching or learning cardiac embryology identified no papers that examined this specific question, and next, a systematic search of YouTube was performed. A total of 1200 cardiac embryology videos were retrieved using 12 specific search terms, with 370 videos retrieved under two or more search terms and excluded. A further 511 videos were excluded under additional, specific criteria. The remaining 319 videos were evaluated with the YouTube Video Assessment Criteria (UTvAC), with 121 rated as "useful." Videos on YouTube are uploaded with a wide audience in mind, from children to cardiologists, and content control is imperfect. Multiple videos were identified as duplicates of videos from original channels, typically without attribution. While 49 videos showed operations or human material, none contained an ethical statement regarding consent, and only 10 of these included an age restriction or graphical advisory. While there are useful videos for medical students studying cardiac embryology on YouTube, intuitive search strategies will also identify many with irrelevant content and of variable quality. Digital competence and search strategies are not innate skills, so educators should teach students to assess information so as to avoid overload or "filter failure."

Visual Aesthetic Quality of Qianjiangyuan National Park Landscapes and Its Spatial Pattern Characteristics

This paper conducts a scientific assessment of aesthetic quality to provide intuitive and scientific planning strategies for national park construction. Focusing on Qianjiangyuan National Park, the study used the scenic beauty evaluation (SBE) method to subjectively assess landscape photos from 16 sample sites. Objective eye movement indicators describing visual behavior were also analyzed. A national park landscape visual quality assessment model was derived through multiple linear regressions correlating subjective evaluations with objective indicators. Spatial technologies like ArcGIS were used to analyze the visual quality and its spatial distribution. Key findings include (1) subjective evaluations showed higher SBE scores for wetland landscapes, followed by recreational, village, and forest landscapes, (2) eye movement behavior varied across landscape types, with the forest landscape having the shortest first fixation time and the lowest saccade frequency, while recreational landscapes had the lowest average saccade speed, (3) strong correlations were found between SBE and indicators such as average fixation time and saccade frequency, with fixation duration ratio being the leading factor influencing visual aesthetic quality, and (4) visual aesthetic quality was highest in the north and south areas of the park, with significant differences between sample sites in these regions compared to the central area. Among different functional zones, the ecological protection area had the highest quality, while the Suzhuang management area excelled in aesthetic quality compared to the Hetian management area.

Post-Truth and Information Warfare in their Technological Context

As citizens are faced with an overabundance of information, their reliance on intuitive sorting strategies and platform-enabled content selection and delivery increases correspondingly. Under such circumstances, political action tends to be based on haphazard encounters with opinion-congruent content than on anything else, giving rise to so-called post-truth condition and, in turn, opening up conditions for manipulating such information encounters as part of information warfare operations. In particular, this novel environment necessitates a rethinking of informational agency, locating it within interactions between humans and technological artefacts, whereby humans as generators of data and algorithms as tools that structure the information domain based on such data co-construct political and social spaces. The impact of digital technologies is further amplified by the advent of synthetic (Artificial Intelligence-generated) media, which is foreseen to bring about epistemic confusion, that is, increasing inability to separate between reality and fiction. Under such conditions, and in any situations of actual or perceived crisis and tension, audiences are inclined to rely on narratives as coping strategies, which is where information warfare operations come to the fore. Either capitalising on the existing fertile ground or having manufactured a condition of crisis and distrust, such operations are geared towards hijacking audience cognitive processes with narratives that suit their perpetrators.

Leadwise clustering multi-branch network for multi-label ECG classification

The 12-lead electrocardiogram (ECG) is widely used for diagnosing cardiovascular diseases in clinical practice. Recently, deep learning methods have become increasingly effective for automatically classifying ECG signals. However, most current research simply combines the 12-lead ECG signals into a matrix without fully considering the intrinsic relationships between the leads and the heart's structure. To better utilize medical domain knowledge, we propose a multi-branch network for multi-label ECG classification and introduce an intuitive and effective lead grouping strategy. Correspondingly, we design multi-branch networks where each branch employs a multi-scale convolutional network structure to extract more comprehensive features, with each branch corresponding to a lead combination. To better integrate features from different leads, we propose a feature weighting fusion module. We evaluate our method on the PTB-XL dataset for classifying 4 arrhythmia types and normal rhythm, and on the China Physiological Signal Challenge 2018 (CPSC2018) database for classifying 8 arrhythmia types and normal rhythm. Experimental results on multiple multi-label datasets demonstrate that our proposed multi-branch network outperforms state-of-the-art networks in multi-label classification tasks

Quantitative RUBY reporter assay for gene regulation analysis.

Various reporter genes have been developed to study gene expression pattern and gene regulation. The RUBY reporter gene was recently developed and widely used, because of its visible and noninvasive advantages. However, quantitative analysis of RUBY gene expression levels was lacking. In this study, we introduce a novel betalain quantification method in combination with the tobacco transient expression system. The betalain produced in tobacco leaves was extracted and purified, and its concentration was quantitatively measured. We successfully applied this approach in studying the transcriptional regulation of ARC5 gene by transcription factors CPD25 and CPD45. Furthermore, with this method, we showed that the gene expression of RCA and Rbcs1A gene were regulated by light, transcription factors HY5 and PIFs through G-box and I-box elements. The development of this betalain quantification approach with the tobacco transient expression system offers a cost-effective and intuitive strategy for studying the regulatory mechanism of gene expression.

A concise but high-performing network for image guided depth completion in autonomous driving

Depth completion is a crucial task in autonomous driving, aiming to convert a sparse depth map into a dense depth prediction. The sparse depth map serves as a partial reference for the actual depth, and the fusion of RGB images is frequently employed to augment the completion process owing to its inherent richness in semantic information. Image-guided depth completion confronts three principal challenges: (1) the effective fusion of the two modalities; (2) the enhancement of depth information recovery; and (3) the realization of real-time predictive capabilities requisite for practical autonomous driving scenarios. In response to these challenges, we propose a concise but high-performing network, named CHNet, to achieve high-performance depth completion with an elegant and straightforward architecture. Firstly, we use a fast guidance module to fuse the two sensor features, harnessing abundant auxiliary information derived from the color space. Unlike the prevalent complex guidance modules, our approach adopts an intuitive and cost-effective strategy. In addition, we find and analyze the optimization inconsistency problem for observed and unobserved positions. To mitigate this challenge, we introduce a decoupled depth prediction head, tailored to better discern and predict depth values for both valid and invalid positions, incurring minimal additional inference time. Capitalizing on the dual-encoder and single-decoder architecture, the simplicity of CHNet facilitates an optimal balance between accuracy and computational efficiency. In benchmark evaluations on the KITTI depth completion dataset, CHNet demonstrates competitive performance metrics and inference speeds relative to contemporary state-of-the-art methodologies. To assess the generalizability of our approach, we extend our evaluations to the indoor NYUv2 dataset, where CHNet continues to yield impressive outcomes. The code of this work will be available at https://github.com/lmomoy/CHNet.

Deep-Representation-Learning-Based Classification Strategy for Anticancer Peptides

Cancer, with its complexity and numerous origins, continues to provide a huge challenge in medical research. Anticancer peptides are a potential treatment option, but identifying and synthesizing them on a large scale requires accurate prediction algorithms. This study presents an intuitive classification strategy, named ACP-LSE, based on representation learning, specifically, a deep latent-space encoding scheme. ACP-LSE can demonstrate notable advancements in classification outcomes, particularly in scenarios with limited sample sizes and abundant features. ACP-LSE differs from typical black-box approaches by focusing on representation learning. Utilizing an auto-encoder-inspired network, it embeds high-dimensional features, such as the composition of g-spaced amino acid pairs, into a compressed latent space. In contrast to conventional auto-encoders, ACP-LSE ensures that the learned feature set is both small and effective for classification, giving a transparent alternative. The suggested approach is tested on benchmark datasets and demonstrates higher performance compared to the current methods. The results indicate improved Matthew’s correlation coefficient and balanced accuracy, offering insights into crucial aspects for developing new ACPs. The implementation of the proposed ACP-LSE approach is accessible online, providing a valuable and reproducible resource for researchers in the field.

Implementations of efficient non-blocking crossbar network with novel multiplexer design in quantum-dot cellular automata

As an emerging nanotechnology, Quantum-dot Cellular Automata (QCA) has attracted extensive attentions due to its characteristics of high density, high speed, and low energy consumption. Modern high-performance System-on-Chips (SoCs) with multiple processors require interconnection networks to connect each core for improving data throughput and reducing latency, while the crossbar network is broadly used as a non-blocking interconnection architecture with high efficiency. In this paper, a method to design an efficient QCA-based N × N crossbar network utilizing optimally designed N:1 multiplexers (MUXs) is proposed, followed by a multi-layer and a single-layer implementation of its 8 × 8 design. A simplified matrix model is then introduced to provide a concise and intuitive switch control strategy, and expressions for cell count, area, latency, QCA cost, and QCA complexity of the proposed crossbar networks are derived according to the size N. Experimental results manifest that the proposed 8 × 8 crossbar networks have significant advantages on most performance parameters compared with other existing QCA-based networks.

Leveraging Domain Expertise in Machine Learning for Critical Metal Prospecting in the Oslo Rift: A Case Study for Fe-Ti-P-Rare Earth Element Mineralization

Global demand for critical raw materials, including phosphorus (P) and rare earth elements (REEs), is on the rise. The south part of Norway, with a particular focus on the Southern Oslo Rift region, is a promising reservoir of Fe-Ti-P-REE resources associated with magmatic systems. Confronting challenges in mineral exploration within these systems, notably the absence of alteration haloes and distal footprints, we have explored alternative methodologies. In this study, we combine machine learning with geological expertise, aiming to identify prospective areas for critical metal prospecting. Our workflow involves processing over 400 rock samples to create training datasets for mineralization and non-mineralization, employing an intuitive sampling strategy to overcome an imbalanced sample ratio. Additionally, we convert airborne magnetic, radiometric, and topographic maps into machine learning-friendly features, with a keen focus on incorporating domain knowledge into these data preparations. Within a binary classification framework, we evaluate two commonly used classifiers: a random forest (RF) and support vector machine (SVM). Our analysis shows that the RF model outperforms the SVM model. The RF model generates a predictive map, identifying approximately 0.3% of the study area as promising for mineralization. These findings align with legacy data and field visits, supporting the map’s potential to guide future surveys.

Occluded Person Re-identification via Saliency-Guided Patch Transfer

While generic person re-identification has made remarkable improvement in recent years, these methods are designed under the assumption that the entire body of the person is available. This assumption brings about a significant performance degradation when suffering from occlusion caused by various obstacles in real-world applications. To address this issue, data-driven strategies have emerged to enhance the model's robustness to occlusion. Following the random erasing paradigm, these strategies typically employ randomly generated noise to supersede randomly selected image regions to simulate obstacles. However, the random strategy is not sensitive to location and content, meaning they cannot mimic real-world occlusion cases in application scenarios. To overcome this limitation and fully exploit the real scene information in datasets, this paper proposes a more intuitive and effective data-driven strategy named Saliency-Guided Patch Transfer (SPT). Combined with the vision transformer, SPT divides person instances and background obstacles using salient patch selection. By transferring person instances to different background obstacles, SPT can easily generate photo-realistic occluded samples. Furthermore, we propose an occlusion-aware Intersection over Union (OIoU) with mask-rolling to filter the more suitable combination and a class-ignoring strategy to achieve more stable processing. Extensive experimental evaluations conducted on occluded and holistic person re-identification benchmarks demonstrate that SPT provides a significant performance gain among different ViT-based ReID algorithms on occluded ReID.

Simulation for Endovascular Treatment

With the advent of high-resolution imaging and advancements in computational fluid dynamics(CFD)and computational structural mechanics(CSM)analyses, clinical simulation of endovascular intervention has gradually become feasible. Virtual stents have become indispensable for coil embolization. For braided stents, such as those with low-profile visualized intraluminal support and flow diverters, predicting postplacement elongation and contraction is challenging; however, software development has enabled more precise treatment planning. Additionally, simulations utilizing three-dimensional(3D)printer models can enable realistic simulations of procedures such as intracranial stents and Woven EndoBridge placement. This approach is beneficial for shunt disorders such as arteriovenous malformations and dural arteriovenous fistulas, offering 3D visualization of shunt access routes and intuitive treatment strategy planning, even for beginners. Furthermore, it can be applied to procedures such as open surgical clipping and nidus resection, aiding in the selection of suitable clips and considerations for ideal resection based on nidus curvature. Simulations using CFD, CSM, and 3D printers are crucial for training surgeons and handling new devices. Harnessing medicine-engineering synergy is essential, and regulatory approval(insurance coverage)and appropriate commercialization of simulations are paramount for the future.

Towards unlocking motor control in spinal cord injured by applying an online EEG-based framework to decode motor intention, trajectory and error processing

Brain-computer interfaces (BCIs) can translate brain signals directly into commands for external devices. Electroencephalography (EEG)-based BCIs mostly rely on the classification of discrete mental states, leading to unintuitive control. The ERC-funded project "Feel Your Reach" aimed to establish a novel framework based on continuous decoding of hand/arm movement intention, for a more natural and intuitive control. Over the years, we investigated various aspects of natural control, however, the individual components had not yet been integrated. Here, we present a first implementation of the framework in a comprehensive online study, combining (i) goal-directed movement intention, (ii) trajectory decoding, and (iii) error processing in a unique closed-loop control paradigm. Testing involved twelve able-bodied volunteers, performing attempted movements, and one spinal cord injured (SCI) participant. Similar movement-related cortical potentials and error potentials to previous studies were revealed, and the attempted movement trajectories were overall reconstructed. Source analysis confirmed the involvement of sensorimotor and posterior parietal areas for goal-directed movement intention and trajectory decoding. The increased experiment complexity and duration led to a decreased performance than each single BCI. Nevertheless, the study contributes to understanding natural motor control, providing insights for more intuitive strategies for individuals with motor impairments.

Flow Allocation Games

We study a game-theoretic variant of the maximum circulation problem. In a flow allocation game, we are given a directed flow network. Each node is a rational agent and can strategically allocate any incoming flow to the outgoing edges. Given the strategy choices of all agents, a maximal circulation that adheres to the chosen allocation strategies evolves in the network. Each agent wants to maximize the amount of flow through his or her node. Flow allocation games can be used to express strategic incentives of clearing in financial networks. We provide a cumulative set of results on the existence and computational complexity of pure Nash and strong equilibria as well as tight bounds on the (strong) prices of anarchy and stability. Our results show an interesting dichotomy. Ranking strategies over individual flow units allows us to obtain optimal strong equilibria for many objective functions. In contrast, more intuitive ranking strategies over edges can give rise to unfavorable incentive properties. Funding: This work was supported by Deutsche Forschungsgemeinschaft Research Group ADYN [411362735].

Additive manufacturing of ULTEM 9085: Weak interface-enriched multi-toughening mechanisms and fracture resistance optimization

Modulating the fracture toughness can be used to tailor or enhance the mechanical performances in the design of specific additively-manufactured engineering structures. In this research, we explored the fracture behavior of ULTEM 9085 materials fabricated by material extrusion (MEX) 3D printing technique, with a particular emphasis on the effect of weak interfaces between deposited beads. Under quasi-static bending condition, the MEX-fabricated ULTEM 9085 exhibits brittle and quasi-brittle fracture behaviors by adjusting the print orientations and raster angles. Several intrinsic toughening mechanisms are identified, including crack deflection and twisting, fibril bridging, constrained microcracking and delamination of the weak interface. Notably, the samples that experienced delamination along the weak interface displayed the highest performance in terms of the fracture toughness. To achieve high fracture toughness, we employed an analytical model based on linear elastic multilayers fracture mechanics to understand the crack onset and predict the crack propagate orientation. Based on the insight gained from this model, we proposed an intuitive optimal strategy to enhance the fracture resistance by fine turning the bonding strength and layer height to trigger the crack deflecting towards and propagating at the weak interface consistently. By implanting this approach, we achieved an increase of over 35% in fracture resistance while effectively preventing the catastrophic failure, demonstrating exceptional fracture-resistant performance. Overall, our study elucidates the impact of layer-by-layer MEX-induced weak interfaces on the fracture resistance of MEX-fabricated high temperature engineering materials. The findings of this research hold potential for engineering applications, particularly in optimizing the design of additive manufactured products aiming at high fracture resistant.

A new adaptive identification strategy of best crop management with farmers

Identification of best performing fertilizer practices with on-farm trials is challenging, in particular in rainfed farming due to weather uncertainty. However, it remains crucial to test a range of viable practices to ascertain their performances, given that they are not known beforehand. This process also involves the testing of practices that could potentially yield inferior results in comparison to the best available practice(s). To identify a best management practice, an “intuitive strategy” typically sets up multi-year, multi-location field trials, wherein each practice is tested in a proportionally equal manner over a set number of years. Our objective was to provide an identification strategy for nitrogen fertilizer management designing a bandit learning algorithm. We aimed for the bandit algorithm to be better at minimizing farmers’ losses occurring from the testing of management practices that do not perform best, compared with the “intuitive strategy” that was formulized as the Explore-Then-Commit strategy. Our case study was for maize production in southern Mali. Bandit framework is a machine learning approach in which an agent learns from the feedback over time and accordingly selects actions in order to maximize its cumulative reward in the long term. To mimic the maize responses to nitrogen fertilization, we used the Decision Support System for Agrotechnology Transfer (DSSAT) crop model. We compared nitrogen fertilizer practices using a risk-aware measure, the Conditional Value-at-Risk (CVaR), and a novel agronomic metric, the Yield Excess (YE). The YE accounts for both grain yield and agronomic nitrogen use efficiency. The bandit algorithm performed better than the intuitive strategy: it minimized farmers’ yield losses during the identification process. This study is a methodological step which opens up new horizons for risk-aware identification of the performance of a range of crop management practices in real conditions.