Data-driven methods have shown tremendous progress in medical image analysis. In this context, deep learning-based supervised methods are widely popular. However, they require a large amount of training data and face issues in generalisability to unseen datasets that hinder clinical translation. Endoscopic imaging data incorporates large inter- and intra-patient variability that makes these models more challenging to learn representative features for downstream tasks. Thus, despite the publicly available datasets and datasets that can be generated within hospitals, most supervised models still underperform. While self-supervised learning has addressed this problem to some extent in natural scene data, there is a considerable performance gap in the medical image domain. In this paper, we propose to explore patch-level instance-group discrimination and penalisation of inter-class variation using additive angular margin within the cosine similarity metrics. Our novel approach enables models to learn to cluster similar representative patches, thereby improving their ability to provide better separation between different classes. Our results demonstrate significant improvement on all metrics over the state-of-the-art (SOTA) methods on the test set from the same and diverse datasets. We evaluated our approach for classification, detection, and segmentation. SSL-CPCD achieves 79.77% on Top 1 accuracy for ulcerative colitis classification, 88.62% on mAP for polyp detection, and 82.32% on dice similarity coefficient for segmentation tasks are nearly over 4%, 2%, and 3%, respectively, compared to the baseline architectures. We also demonstrate that our method generalises better than all SOTA methods to unseen datasets, reporting nearly 7% improvement in our generalisability assessment.
Deep learning could be prone to learning shortcuts raised by dataset bias and result in inaccurate, unreliable, and unfair models, which impedes its adoption in real-world clinical applications. Despite its significance, there is a dearth of research in the medical image classification domain to address dataset bias. Furthermore, the bias labels are often agnostic, as identifying biases can be laborious and depend on post-hoc interpretation. This paper proposes learning Adaptive Agreement from a Biased Council (Ada-ABC), a debiasing framework that does not rely on explicit bias labels to tackle dataset bias in medical images. Ada-ABC develops a biased council consisting of multiple classifiers optimized with generalized cross entropy loss to learn the dataset bias. A debiasing model is then simultaneously trained under the guidance of the biased council. Specifically, the debiasing model is required to learn adaptive agreement with the biased council by agreeing on the correctly predicted samples and disagreeing on the wrongly predicted samples by the biased council. In this way, the debiasing model could learn the target attribute on the samples without spurious correlations while also avoiding ignoring the rich information in samples with spurious correlations. We theoretically demonstrated that the debiasing model could learn the target features when the biased model successfully captures dataset bias. Moreover, to our best knowledge, we constructed the first medical debiasing benchmark from four datasets containing seven different bias scenarios. Our extensive experiments practically showed that our proposed Ada-ABC outperformed competitive approaches, verifying its effectiveness in mitigating dataset bias for medical image classification. The codes and organized benchmark datasets will be made publicly available.
연합 학습(Federated Learning, FL)은 데이터를 중앙 집중화하지 않고도 분산된 이미지 저장소 전반에서 심층 학습(deep learning) 모델의 협력적 학습(training)을 가능하게 한다. 이러한 패러다임은 법적 규제, 개인정보 보호 우려, 대역폭 제약으로 인해 데이터 공유가 제한되는 원격 탐사(Remote Sensing, RS) 분야에서 특히 유용하다. 그러나 클라이언트 간 학습 데이터의 이질성(non-IID 데이터로 알려짐)은 수렴(convergence)을 저해하고 집계된 글로벌 모델의 일반화 능력을 제한할 수 있다. 학습 데이터 이질성의 부정적인 영향을 완화하기 위해, 분포 변화(distribution shift) 하에서 강건성을 입증한 전이 가능한 표현력을 지닌 비전-언어 모델(Vision-Language Model, VLM)을 연합 학습에 활용할 수 있다. 그러나 큰 매개변수(parameter) 크기는 연합 환경에서 통신 오버헤드와 로컬 계산 복잡도를 크게 증가시킬 수 있다. 따라서 일반화 능력과 통신 및 계산 제약 사이의 균형을 맞추는 적절한 VLM 적응 전략을 선택하는 것이 매우 중요하다. 이러한 문제를 해결하기 위해, 본 논문에서는 RS 이미지 분류(image classification) 관점에서 연합 학습을 위한 VLM 적응 전략에 대한 최초의 비교 연구를 제시한다. 전체 미세조정(full fine-tuning), 인코더별 미세조정, 프롬프트(prompt) 학습, 그리고 LoRA(Low-Rank Adaptation) 튜닝을 조사하고 1) non-IID 데이터 하에서의 일반화 능력, 2) 통신 오버헤드, 3) 로컬 계산 복잡도라는 세 가지 기준에 따라 이를 분석한다. BigEarthNet-S2, EuroSAT, RESISC45 및 ImageNet에서의 실험은 태스크 특화, 교차 도메인 일반화, 효율성 사이의 뚜렷한 트레이드오프를 보여준다. 이러한 결과를 바탕으로, 다양한 운영 제약 하에서 RS 이미지 분류를 위한 연합 학습 시 적절한 VLM 적응 전략을 선택할 수 있는 가이드라인을 도출한다. 본 연구의 코드는 https://git.tu-berlin.de/rsim/FL-RS-VLM 에서 공개되어 있다.
비전-언어 모델과 생성형 편집 시스템의 최근 발전으로 이미지, 텍스트, 또는 둘을 함께 바꿔 설득력 있는 멀티모달 허위정보를 만들기가 점점 쉬워졌다. 그러나 기존 데이터셋은 주로 진위(authenticity), 문맥 외 불일치(out-of-context mismatch), 또는 조작 유형에 초점을 맞추며, 편집이 게시물의 가능한 해석을 얼마나 강하게 바꾸는지는 거의 포착하지 못한다.
T-IMPACT는 조작된 뉴스 스타일 이미지-텍스트 쌍을 위한 첫 공개 심각도 인식 벤치마크이다. T-IMPACT는 원본(pristine), 이미지만, 텍스트만, 및 공동(joint) 조작을 포괄하는 98,786개 예시를 포함하며, 보정된 연속 심각도 신호, 거친 저/중/고 라벨, 그리고 그라운딩(grounding) 메타데이터를 제공한다.
뉴스 이미지-텍스트 쌍에서 출발하여, 파이프라인은 의미적 앵커(semantic anchor)를 추출하고 이를 공간적으로 그라운딩(grounding)한 뒤, 국소적 이미지 편집과 제약된 캡션 재작성을 수행하고, 제한된 인간 평가를 이용해 문맥적 영향(contextual impact) 점수를 보정한다.
이번 릴리스에서는 보정된 연속 점수가 주요 심각도 목표이며, 저/중/고 구간은 균형 잡힌 클래스가 아니라 거친 운용 구간(coarse operating bucket)으로 해석해야 한다.
실험 결과, 현재 모델은 일부 진위 신호를 회복하지만, 심각도 예측은 여전히 훨씬 더 어렵고 인간 판단과는 약하게만 정렬(alignment)된다. T-IMPACT는 이진 진짜/가짜(real/fake) 분류를 넘어 등급화된 문맥적 영향을 향한 멀티모달 조작 연구를 위한 초기 벤치마크를 제공한다.
머신러닝 모델의 예측을 해석하는 것은 특히 블랙박스 비전 모델에서 핵심적인 과제입니다. 개념 기반 기여도 분석(concept-based attribution) 방법은 사람이 이해할 수 있는 개념을 사용하여 모델을 해석합니다. 기존의 개념 기반 방법들은 분류(classification) 및 회귀 모델만을 해석할 수 있습니다. 또한 모델 가중치에 대한 접근이나 대규모 개념 주석 데이터셋을 필요로 하며, 대부분의 방법은 이미지 수준의 기여도를 제공하지 못합니다. 본 논문에서는 모델 가중치 접근이나 라벨이 지정된 개념 데이터셋 없이도 동작하는 새로운 개념 수준 기여도 분석 방법인 Hide and Seek(HnS)를 제안합니다. HnS는 광범위한 비전 모델을 해석할 수 있습니다. 개념과 이미지 세트가 주어지면, HnS는 비전-언어 모델(VLM)을 사용하여 개념과 일치하는 이미지를 검색하고, 각 이미지 내에서 해당 개념의 위치를 특정(localize)한 뒤, 해당 영역을 마스킹하고, 원본 이미지와 마스킹된 이미지 간의 블랙박스 모델 출력 변화로부터 기여도를 측정합니다. 이를 통해 모델의 동작을 직접 측정하고 국소적(local) 기여도를 지원함으로써 충실한 기여도 분석을 가능하게 합니다. 저자들은 CelebA와 COCO-Stuff에서 HnS를 검증했습니다. 기존 방법들과 비교했을 때, HnS는 정답(ground-truth) 기여도와 더 높은 일치도를 보였습니다. 또한 기존의 개념 기여도 분석 방법을 적용할 수 없었던 이미지 캡셔닝 및 위성 이미지 분할(image segmentation)을 포함한 다양한 태스크에서 HnS의 적용 가능성을 입증했습니다. 익명화된 코드는 보충 자료(supplementary material)에 제공되었습니다.
Image classification problems are typically addressed by first collecting examples with candidate labels, second cleaning the candidate labels manually, and third training a deep neural network on the clean examples. The manual labeling step is often the most expensive one as it requires workers to label millions of images. In this paper we propose to work without any explicitly labeled data by i) directly training the deep neural network on the noisy candidate labels, and ii) early stopping the training to avoid overfitting. With this procedure we exploit an intriguing property of standard overparameterized convolutional neural networks trained with (stochastic) gradient descent: Clean labels are fitted faster than noisy ones. We consider two classification problems, a subset of ImageNet and CIFAR-10. For both, we construct large candidate datasets without any explicit human annotations, that only contain 10%-50% correctly labeled examples per class. We show that training on the candidate examples and regularizing through early stopping gives higher test performance for both problems than when training on the original, clean data. This is possible because the candidate datasets contain a huge number of clean examples, and, as we show in this paper, the noise generated through the label collection process is not nearly as adversarial for learning as the noise generated by randomly flipping labels.
Deep learning systems are known to be vulnerable to adversarial examples. In particular, query-based black-box attacks do not require knowledge of the deep learning model, but can compute adversarial examples over the network by submitting queries and inspecting returns. Recent work largely improves the efficiency of those attacks, demonstrating their practicality on today's ML-as-a-service platforms. We propose Blacklight, a new defense against query-based black-box adversarial attacks. The fundamental insight driving our design is that, to compute adversarial examples, these attacks perform iterative optimization over the network, producing image queries highly similar in the input space. Blacklight detects query-based black-box attacks by detecting highly similar queries, using an efficient similarity engine operating on probabilistic content fingerprints. We evaluate Blacklight against eight state-of-the-art attacks, across a variety of models and image classification tasks. Blacklight identifies them all, often after only a handful of queries. By rejecting all detected queries, Blacklight prevents any attack to complete, even when attackers persist to submit queries after account ban or query rejection. Blacklight is also robust against several powerful countermeasures, including an optimal black-box attack that approximates white-box attacks in efficiency. Finally, we illustrate how Blacklight generalizes to other domains like text classification.
There is a warning light for the loss of plant habitats worldwide that entails concerted efforts to conserve plant biodiversity. Thus, plant species classification is of crucial importance to address this environmental challenge. In recent years, there is a considerable increase in the number of studies related to plant taxonomy. While some researchers try to improve their recognition performance using novel approaches, others concentrate on computational optimization for their framework. In addition, a few studies are diving into feature extraction to gain significantly in terms of accuracy. In this paper, we propose an effective method for the leaf recognition problem. In our proposed approach, a leaf goes through some pre-processing to extract its refined color image, vein image, xy-projection histogram, handcrafted shape, texture features, and Fourier descriptors. These attributes are then transformed into a better representation by neural network-based encoders before a support vector machine (SVM) model is utilized to classify different leaves. Overall, our approach performs a state-of-the-art result on the Flavia leaf dataset, achieving the accuracy of 99.58\% on test sets under random 10-fold cross-validation and bypassing the previous methods. We also release our codes (Scripts are available at https://github.com/dinhvietcuong1996/LeafRecognition) for contributing to the research community in the leaf classification problem.
The accurate analysis of medical images is vital for diagnosing and predicting medical conditions. Traditional approaches relying on radiologists and clinicians suffer from inconsistencies and missed diagnoses. Computer-aided diagnosis systems can assist in achieving early, accurate, and efficient diagnoses. This paper presents an improved Evolutionary Algorithm-based Transformer architecture for medical image classification using Vision Transformers. The proposed EATFormer architecture combines the strengths of Convolutional Neural Networks and Vision Transformers, leveraging their ability to identify patterns in data and adapt to specific characteristics. The architecture incorporates novel components, including the Enhanced EA-based Transformer block with Feed-Forward Network, Global and Local Interaction , and Multi-Scale Region Aggregation modules. It also introduces the Modulated Deformable MSA module for dynamic modeling of irregular locations. The paper discusses the Vision Transformer (ViT) model's key features, such as patch-based processing, positional context incorporation, and Multi-Head Attention mechanism. It introduces the Multi-Scale Region Aggregation module, which aggregates information from different receptive fields to provide an inductive bias. The Global and Local Interaction module enhances the MSA-based global module by introducing a local path for extracting discriminative local information. Experimental results on the Chest X-ray and Kvasir datasets demonstrate that the proposed EATFormer significantly improves prediction speed and accuracy compared to baseline models.
Continual learning (CL) breaks off the one-way training manner and enables a model to adapt to new data, semantics and tasks continuously. However, current CL methods mainly focus on single tasks. Besides, CL models are plagued by catastrophic forgetting and semantic drift since the lack of old data, which often occurs in remote-sensing interpretation due to the intricate fine-grained semantics. In this paper, we propose Continual Panoptic Perception (CPP), a unified continual learning model that leverages multi-task joint learning covering pixel-level classification, instance-level segmentation and image-level perception for universal interpretation in remote sensing images. Concretely, we propose a collaborative cross-modal encoder (CCE) to extract the input image features, which supports pixel classification and caption generation synchronously. To inherit the knowledge from the old model without exemplar memory, we propose a task-interactive knowledge distillation (TKD) method, which leverages cross-modal optimization and task-asymmetric pseudo-labeling (TPL) to alleviate catastrophic forgetting. Furthermore, we also propose a joint optimization mechanism to achieve end-to-end multi-modal panoptic perception. Experimental results on the fine-grained panoptic perception dataset validate the effectiveness of the proposed model, and also prove that joint optimization can boost sub-task CL efficiency with over 13\% relative improvement on panoptic quality.
자기공명영상(MRI)의 자동 분류는 신경계 질환을 비침습적으로 진단하는 중요한 접근법이다. 그러나 촬영 과정에서 생기는 잡음과 사람마다 구조가 서로 비슷하다는 점 때문에 미세한 병리 차이가 가려져 심층 신경망의 성능이 제한된다. 대조 학습(CL)은 이 문제를 완화할 가능성이 있지만 기존 방법은 경직된 이진 구분을 사용해 음성 표본의 이질성을 무시한다. 또한 세밀한 해부학적 정렬에 중요한 시점 간 구조적 일관성을 포착하지 못한다. 본 연구는 구조적 일관성과 의미적 구별 능력을 함께 높이도록 설계한 이중 분기 계층적 대조 학습(DBHiCo)을 제안한다. 원본 MRI와 두개골 등 뇌 바깥 조직을 제거한 MRI를 자연스러운 해부학적 다중 관점의 쌍으로 사용한다. 이중 분기 아키텍처에서 처리해 전체 구조의 강건성과 두개강 내부의 국소 세부 정보를 통합한다. 핵심은 기존 이진 제약을 마진으로 순위를 정한 유사도 계층으로 대체하는 계층적 대조 손실이다. 구조 및 의미 측면의 난이도에 따라 양성 쌍과 음성 쌍에 명시적으로 순서를 부여해 학습한 표현의 세밀한 의미 판별력을 높인다. 서로 다른 신경계 데이터셋 세 종의 광범위한 실험에서 DBHiCo는 최신 최고 성능을 달성했다. 기존 MRI 다중 관점 표현 학습과 최신 대조 미세조정 방법보다 일관되게 성능이 높았다.
컴퓨터 병리학에서 전체 슬라이드 이미지(WSI) 분석은 패치 임베딩을 독립적으로 추출하고 슬라이드 수준 예측을 위해 집계하는 다중 인스턴스 학습(MIL) 파이프라인을 따르지만, 염색, 스캐너 및 국소적 질감으로 인한 슬라이드 내 분산이 판별 신호를 압도할 수 있습니다. 본 논문에서는 경량 MLP 어댑터를 윈도우 기반 마스킹 재구성 브랜치 및 MIL 분류(classification) 헤드와 함께 공동으로 학습시키는 특징 공간 마스킹 프레임워크인 MFE-MIL(Masked Feature Encoding for Multiple Instance Learning)을 제안합니다. 이 두 목적 함수는 상호 보완적입니다. 분류는 어댑터가 슬라이드 내 패치 분산을 억제하도록 유도하는 반면, 윈도우 기반 마스킹 재구성은 패치 좌표, 좌표 그래프 또는 분할 전처리 없이 적응된 특징(feature)에 대한 보조 정규화 역할을 합니다. 래스터 패치 추출 순서는 약한 암시적 사전 정보로만 사용됩니다. 추론(inference) 시에는 디코더가 제거되어 어댑터와 MIL 헤드만 남습니다. 4개의 서로 다른 인코더를 사용한 CAMELYON16/17, PANDA, TCGA-BRCA 전반에서 MFE-MIL은 테스트된 거의 모든 집계기-인코더 설정에 대해 ACC/F1을 향상시키고 대부분에서 AUC를 개선하며, 좌표 기반 공간 방식(CAMIL)을 능가하고, 4개 데이터셋 중 3개(UNI)에서 2DMamba보다 높은 AUC를 달성합니다. 5개의 TCGA 생존 코호트에서는 테스트된 모든 집계기에 대해 평균 일치도 지수(concordance index)를 개선하여 가장 일관된 성능 향상을 보였습니다. 코드는 https://github.com/AtlasAnalyticsLab/MFE-MIL 에서 확인할 수 있습니다.
Image captioning implies automatically generating textual descriptions of images based only on the visual input. Although this has been an extensively addressed research topic in recent years, not many contributions have been made in the domain of art historical data. In this particular context, the task of image captioning is confronted with various challenges such as the lack of large-scale datasets of image-text pairs, the complexity of meaning associated with describing artworks and the need for expert-level annotations. This work aims to address some of those challenges by utilizing a novel large-scale dataset of artwork images annotated with concepts from the Iconclass classification system designed for art and iconography. The annotations are processed into clean textual description to create a dataset suitable for training a deep neural network model on the image captioning task. Motivated by the state-of-the-art results achieved in generating captions for natural images, a transformer-based vision-language pre-trained model is fine-tuned using the artwork image dataset. Quantitative evaluation of the results is performed using standard image captioning metrics. The quality of the generated captions and the model's capacity to generalize to new data is explored by employing the model on a new collection of paintings and performing an analysis of the relation between commonly generated captions and the artistic genre. The overall results suggest that the model can generate meaningful captions that exhibit a stronger relevance to the art historical context, particularly in comparison to captions obtained from models trained only on natural image datasets.
Deep learning is found to be vulnerable to adversarial examples. However, its adversarial susceptibility in image caption generation is under-explored. We study adversarial examples for vision and language models, which typically adopt an encoder-decoder framework consisting of two major components: a Convolutional Neural Network (i.e., CNN) for image feature extraction and a Recurrent Neural Network (RNN) for caption generation. In particular, we investigate attacks on the visual encoder's hidden layer that is fed to the subsequent recurrent network. The existing methods either attack the classification layer of the visual encoder or they back-propagate the gradients from the language model. In contrast, we propose a GAN-based algorithm for crafting adversarial examples for neural image captioning that mimics the internal representation of the CNN such that the resulting deep features of the input image enable a controlled incorrect caption generation through the recurrent network. Our contribution provides new insights for understanding adversarial attacks on vision systems with language component. The proposed method employs two strategies for a comprehensive evaluation. The first examines if a neural image captioning system can be misled to output targeted image captions. The second analyzes the possibility of keywords into the predicted captions. Experiments show that our algorithm can craft effective adversarial images based on the CNN hidden layers to fool captioning framework. Moreover, we discover the proposed attack to be highly transferable. Our work leads to new robustness implications for neural image captioning.
We propose a method to estimate the uncertainty of the outcome of an image classifier on a given input datum. Deep neural networks commonly used for image classification are deterministic maps from an input image to an output class. As such, their outcome on a given datum involves no uncertainty, so we must specify what variability we are referring to when defining, measuring and interpreting uncertainty, and attributing "confidence" to the outcome. To this end, we introduce the Wellington Posterior, which is the distribution of outcomes that would have been obtained in response to data that could have been generated by the same scene that produced the given image. Since there are infinitely many scenes that could have generated any given image, the Wellington Posterior involves inductive transfer from scenes other than the one portrayed. We explore the use of data augmentation, dropout, ensembling, single-view reconstruction, and model linearization to compute a Wellington Posterior. Additional methods include the use of conditional generative models such as generative adversarial networks, neural radiance fields, and conditional prior networks. We test these methods against the empirical posterior obtained by performing inference on multiple images of the same underlying scene. These developments are only a small step towards assessing the reliability of deep network classifiers in a manner that is compatible with safety-critical applications and human interpretation.
The Internet of Things (IoT) is an extension of the traditional Internet, which allows a very large number of smart devices, such as home appliances, network cameras, sensors and controllers to connect to one another to share information and improve user experiences. Current IoT devices are typically micro-computers for domain-specific computations rather than traditional functionspecific embedded devices. Therefore, many existing attacks, targeted at traditional computers connected to the Internet, may also be directed at IoT devices. For example, DDoS attacks have become very common in IoT environments, as these environments currently lack basic security monitoring and protection mechanisms, as shown by the recent Mirai and Brickerbot IoT botnets. In this paper, we propose a novel light-weight approach for detecting DDos malware in IoT environments.We firstly extract one-channel gray-scale images converted from binaries, and then utilize a lightweight convolutional neural network for classifying IoT malware families. The experimental results show that the proposed system can achieve 94.0% accuracy for the classification of goodware and DDoS malware, and 81.8% accuracy for the classification of goodware and two main malware families.
Multispectral image analysis is a relatively promising field of research with applications in several areas, such as medical imaging and satellite monitoring. A considerable number of current methods of analysis are based on parametric statistics. Alternatively, some methods in Computational Intelligence are inspired by biology and other sciences. Here we claim that Philosophy can be also considered as a source of inspiration. This work proposes the Objective Dialectical Method (ODM): a method for classification based on the Philosophy of Praxis. ODM is instrumental in assembling evolvable mathematical tools to analyze multispectral images. In the case study described in this paper, multispectral images are composed of diffusion-weighted (DW) magnetic resonance (MR) images. The results are compared to ground-truth images produced by polynomial networks using a morphological similarity index. The classification results are used to improve the usual analysis of the apparent diffusion coefficient map. Such results proved that gray and white matter can be distinguished in DW-MR multispectral analysis and, consequently, DW-MR images can also be used to furnish anatomical information.
New results suggest strong limits to the feasibility of classifying human brain activity evoked from image stimuli, as measured through EEG. Considerable prior work suffers from a confound between the stimulus class and the time since the start of the experiment. A prior attempt to avoid this confound using randomized trials was unable to achieve results above chance in a statistically significant fashion when the data sets were of the same size as the original experiments. Here, we again attempt to replicate these experiments with randomized trials on a far larger (20x) dataset of 1,000 stimulus presentations of each of forty classes, all from a single subject. To our knowledge, this is the largest such EEG data collection effort from a single subject and is at the bounds of feasibility. We obtain classification accuracy that is marginally above chance and above chance in a statistically significant fashion, and further assess how accuracy depends on the classifier used, the amount of training data used, and the number of classes. Reaching the limits of data collection without substantial improvement in classification accuracy suggests limits to the feasibility of this enterprise.
A relatively new set of transport-based transforms (CDT, R-CDT, LOT) have shown their strength and great potential in various image and data processing tasks such as parametric signal estimation, classification, cancer detection among many others. It is hence worthwhile to elucidate some of the mathematical properties that explain the successes of these transforms when they are used as tools in data analysis, signal processing or data classification. In particular, we give conditions under which classes of signals that are created by algebraic generative models are transformed into convex sets by the transport transforms. Such convexification of the classes simplify the classification and other data analysis and processing problems when viewed in the transform domain. More specifically, we study the extent and limitation of the convexification ability of these transforms under an algebraic generative modeling framework. We hope that this paper will serve as an introduction to these transforms and will encourage mathematicians and other researchers to further explore the theoretical underpinnings and algorithmic tools that will help understand the successes of these transforms and lay the groundwork for further successful applications.
Multiclass classification (MCC) is a fundamental machine learning problem of classifying each instance into one of a predefined set of classes. In the deep learning era, extensive efforts have been spent on developing more powerful neural embedding models to better represent the instance for improving MCC performance. In this paper, we do not aim to propose new neural models for instance representation learning, but to show that it is promising to boost MCC performance with a novel formulation through the lens of ranking. In particular, by viewing MCC as to rank classes for an instance, we first argue that ranking metrics, such as Normalized Discounted Cumulative Gain, can be more informative than the commonly used Top-$K$ metrics. We further demonstrate that the dominant neural MCC recipe can be transformed to a neural ranking framework. Based on such generalization, we show that it is intuitive to leverage advanced techniques from the learning to rank literature to improve the MCC performance out of the box. Extensive empirical results on both text and image classification tasks with diverse datasets and backbone neural models show the value of our proposed framework.