本文提出 Continuous Histogram Loss (CHL) 损失函数，将二元相似性推广至连续多值相似性，通过可微二维直方图计算距离与相似性的条件概率，以解决相似性学习、表示学习和数据可视化等更广泛任务。
摘要：
相似性学习近年来引起了研究人员的广泛关注，最近也提出了大量成功的方法。然而，大多数最先进的相似性学习方法只考虑二元相似性。在本文中，我们引入了一种名为 Continuous Histogram Loss (CHL) 的新损失函数，它将最近提出的 Histogram loss 推广到多值相似性，即允许相似性的可接受值在某个范围内连续分布。该新颖的损失函数通过以可微的方式将成对距离和相似性聚合到二维直方图中，然后计算成对距离不会随着相似性增加而减少这一条件的概率来计算。这种新颖的损失能够解决更广泛的任务，包括相似性学习、表示学习和数据可视化。
英文摘要：
Similarity learning has gained a lot of attention from researches in recent years and tons of successful approaches have been recently proposed. However, the majority of the state-of-the-art similarity learning methods consider only a binary similarity. In this paper we introduce a new loss function called Continuous Histogram Loss (CHL) which generalizes recently proposed Histogram loss to multiple-valued similarities, i.e. allowing the acceptable values of similarity to be continuously distributed within some range. The novel loss function is computed by aggregating pairwise distances and similarities into 2D histograms in a differentiable manner and then computing the probability of condition that pairwise distances will not decrease as the similarities increase. The novel loss is capable of solving a wider range of tasks including similarity learning, representation learning and data visualization.
Ocean University of China（中国海洋大学）；Nanjing University（南京大学）
Image retrieval has been a top topic in the field of both computer vision and machine learning for a long time. Content based image retrieval, which tries to retrieve images from a database visually similar to a query image, has attracted much attention. Two most important issues of image retrieval are the representation and ranking of the images. Recently, bag-of-words based method has shown its power as a representation method. Moreover, nonnegative matrix factorization is also a popular way to represent the data samples. In addition, contextual similarity learning has also been studied and …
Learning the Relation between Similarity Loss and Clustering Loss in Self-Supervised Learning
Nanjing University（南京大学）；Southeast University（东南大学）；Purple Mountain Laboratories（紫金山实验室）；Amazon.com L.L.C.（亚马逊有限责任公司）；Alibaba Group（阿里巴巴集团）；The Hong Kong University of Science and Technology（香港科技大学）；Southern Methodist University（南卫理公会大学）
Sabzevar University of New Technology（萨布泽瓦尔新技术大学）；Ferdowsi University of Mashhad（马什哈德费尔多西大学）；Hakim Sabzevari University（哈基姆萨布泽瓦里大学）
An important challenge in metric learning is scalability to both size and dimension of input data. Online metric learning algorithms are proposed to address this challenge. Existing methods are commonly based on (Passive Aggressive) PA approach. Hence, they can rapidly process large volumes of data with an adaptive learning rate. However, these algorithms are based on the Hinge loss and so are not robust against outliers and label noise. Also, existing online methods usually assume training triplets or pairwise constraints are exist in advance. However, many datasets in real-world applications…
The University of Sydney（悉尼大学）；Sydney Institute for Astronomy（悉尼天文研究所）
Although ubiquitous in the sciences, histogram data have not received much attention by the Deep Learning community. Whilst regression and classification tasks for scalar and vector data are routinely solved by neural networks, a principled approach for estimating histogram labels as a function of an input vector or image is lacking in the literature. We present a dedicated method for Deep Learning-based histogram regression, which incorporates cross-bin information and yields distributions over possible histograms, expressed by $τ$-quantiles of the cumulative histogram in each bin. The crux o…
Self-supervised learning is a growing paradigm in deep representation learning, showing great generalization capabilities and competitive performance in low-labeled data regimes. The SimCLR framework proposes the NT-Xent loss for contrastive representation learning. The objective of the loss function is to maximize agreement, similarity, between sampled positive pairs. This short paper derives and proposes an upper bound for the loss and average similarity. An analysis of the implications is however not provided, but we strongly encourage anyone in the field to conduct this.
Institute of Computer Science and Technology, Peking University（北京大学计算机科学技术研究所）
DNN-based cross-modal retrieval has become a research hotspot, by which users can search results across various modalities like image and text. However, existing methods mainly focus on the pairwise correlation and reconstruction error of labeled data. They ignore the semantically similar and dissimilar constraints between different modalities, and cannot take advantage of unlabeled data. This paper proposes Cross-modal Deep Metric Learning with Multi-task Regularization (CDMLMR), which integrates quadruplet ranking loss and semi-supervised contrastive loss for modeling cross-modal semantic si…
As a highlighting research topic in the multimedia area, cross-media retrieval aims to capture the complex correlations among multiple media types. Learning better shared representation and distance metric for multimedia data is important to boost the cross-media retrieval. Motivated by the strong ability of deep neural network in feature representation and comparison functions learning, we propose the Unified Network for Cross-media Similarity Metric (UNCSM) to associate cross-media shared representation learning with distance metric in a unified framework. First, we design a two-pathway deep…
Alberta Machine Intelligence Institute (Amii) and Reinforcement Learning and Artificial Intelligence Laboratory（阿尔伯塔机器智能研究所（Amii）和强化学习与人工智能实验室）；Department of Computing Science, University of Alberta（计算科学系，阿尔伯塔大学）；University of Tübingen（图宾根大学）；Zuse School ELIZA（祖斯学校ELIZA）
NEC Labs America（美国 NEC 实验室）；University of California San Diego（加利福尼亚大学圣迭戈分校）
Face recognition models trained under the assumption of identical training and test distributions often suffer from poor generalization when faced with unknown variations, such as a novel ethnicity or unpredictable individual make-ups during test time. In this paper, we introduce a novel cross-domain metric learning loss, which we dub Cross-Domain Triplet (CDT) loss, to improve face recognition in unseen domains. The CDT loss encourages learning semantically meaningful features by enforcing compact feature clusters of identities from one domain, where the compactness is measured by underlying …
We present the Hue-Net - a novel Deep Learning framework for Intensity-based Image-to-Image Translation. The key idea is a new technique termed network augmentation which allows a differentiable construction of intensity histograms from images. We further introduce differentiable representations of (1D) cyclic and joint (2D) histograms and use them for defining loss functions based on cyclic Earth Mover's Distance (EMD) and Mutual Information (MI). While the Hue-Net can be applied to several image-to-image translation tasks, we choose to demonstrate its strength on color transfer problems, whe…
Peking University（北京大学）；University of Twente（特文特大学）；Pazhou Lab（琶洲实验室）
In this paper, we propose a density estimation algorithm called \textit{Gradient Boosting Histogram Transform} (GBHT), where we adopt the \textit{Negative Log Likelihood} as the loss function to make the boosting procedure available for the unsupervised tasks. From a learning theory viewpoint, we first prove fast convergence rates for GBHT with the smoothness assumption that the underlying density function lies in the space $C^{0,α}$. Then when the target density function lies in spaces $C^{1,α}$, we present an upper bound for GBHT which is smaller than the lower bound of its corresponding bas…
Washington State University（华盛顿州立大学）；DeepMind（深度思维）；Google Research（谷歌研究院）
One approach to deal with the statistical inefficiency of neural networks is to rely on auxiliary losses that help to build useful representations. However, it is not always trivial to know if an auxiliary task will be helpful for the main task and when it could start hurting. We propose to use the cosine similarity between gradients of tasks as an adaptive weight to detect when an auxiliary loss is helpful to the main loss. We show that our approach is guaranteed to converge to critical points of the main task and demonstrate the practical usefulness of the proposed algorithm in a few domains…
We propose a novel domain specific loss, which is a differentiable loss function based on the dose volume histogram, and combine it with an adversarial loss for the training of deep neural networks to generate Pareto optimal dose distributions. The mean squared error (MSE) loss, dose volume histogram (DVH) loss, and adversarial (ADV) loss were used to train 4 instances of the neural network model: 1) MSE, 2) MSE+ADV, 3) MSE+DVH, and 4) MSE+DVH+ADV. 70 prostate patients were acquired, and the dose influence arrays were calculated for each patient. 1200 Pareto surface plans per patient were gene…
Quantification using Permutation-Invariant Networks based on Histograms
Artificial Intelligence Center（人工智能中心）；University of Oviedo（奥维耶多大学）；Istituto di Scienza e Tecnologie dell’Informazione（信息科学与技术研究所）；Consiglio Nazionale delle Ricerche（国家研究委员会）
University of Pennsylvania（宾夕法尼亚大学）；Montreal Institute for Learning Algorithms（蒙特利尔学习算法研究所）
In this work, we demonstrate that a major limitation of regression using a mean-squared error loss is its sensitivity to the scale of its targets. This makes learning settings consisting of target's whose values take on varying scales challenging. A recently-proposed alternative loss function, known as histogram loss, avoids this issue. However, its computational cost grows linearly with the number of buckets in the histogram, which renders prediction with real-valued targets intractable. To address this issue, we propose a novel approach to training deep learning models on real-valued regress…
Convolution Neural Networks (CNN) have recently achieved state-of-the art performance on handwritten Chinese character recognition (HCCR). However, most of CNN models employ the SoftMax activation function and minimize cross entropy loss, which may cause loss of inter-class information. To cope with this problem, we propose to combine cross entropy with similarity ranking function and use it as loss function. The experiments results show that the combination loss functions produce higher accuracy in HCCR. This report briefly reviews cross entropy loss function, a typical similarity ranking fun…
The Mean Square Error (MSE) has shown its strength when applied in deep generative models such as Auto-Encoders to model reconstruction loss. However, in image domain especially, the limitation of MSE is obvious: it assumes pixel independence and ignores spatial relationships of samples. This contradicts most architectures of Auto-Encoders which use convolutional layers to extract spatial dependent features. We base on the structural similarity metric (SSIM) and propose a novel level weighted structural similarity (LWSSIM) loss for convolutional Auto-Encoders. Experiments on common datasets on…
Class labels used for machine learning are relatable to each other, with certain class labels being more similar to each other than others (e.g. images of cats and dogs are more similar to each other than those of cats and cars). Such similarity among classes is often the cause of poor model performance due to the models confusing between them. Current labeling techniques fail to explicitly capture such similarity information. In this paper, we instead exploit the similarity between classes by capturing the similarity information with our novel confidence labels. Confidence labels are probabil…
Institute for Intelligent Systems（智能系统研究所）；University of Johannesburg（约翰内斯堡大学）
Catastrophic forgetting in neural networks during incremental learning remains a challenging problem. Previous research investigated catastrophic forgetting in fully connected networks, with some earlier work exploring activation functions and learning algorithms. Applications of neural networks have been extended to include similarity learning. Understanding how similarity learning loss functions would be affected by catastrophic forgetting is of significant interest. Our research investigates catastrophic forgetting for four well-known similarity-based loss functions during incremental class…
Hamburg University of Technology（汉堡工业大学）；Imperial College London（帝国理工学院）
Supervised machine learning approaches require the formulation of a loss functional to be minimized in the training phase. Sequential data are ubiquitous across many fields of research, and are often treated with Euclidean distance-based loss functions that were designed for tabular data. For smooth oscillatory data, those conventional approaches lack the ability to penalize amplitude, frequency and phase prediction errors at the same time, and tend to be biased towards amplitude errors. We introduce the surface similarity parameter (SSP) as a novel loss function that is especially useful for …
Technion Israel Institute of Technology（以色列理工学院）；McGovern Institute for Brain Research, Massachusetts Institute of Technology（麻省理工学院麦戈文脑科学研究所）
Compressed sensing (CS) is a signal processing framework for efficiently reconstructing a signal from a small number of measurements, obtained by linear projections of the signal. In this paper we present an end-to-end deep learning approach for CS, in which a fully-connected network performs both the linear sensing and non-linear reconstruction stages. During the training phase, the sensing matrix and the non-linear reconstruction operator are jointly optimized using Structural similarity index (SSIM) as loss rather than the standard Mean Squared Error (MSE) loss. We compare the proposed appr…
Indiana University-Purdue University Indianapolis（印第安纳大学与普渡大学印第安纳波利斯联合分校）；eBay Inc.（易贝公司）
Query similarity prediction task is generally solved by regression based models with square loss. Such a model is agnostic of absolute similarity values and it penalizes the regression error at all ranges of similarity values at the same scale. However, to boost e-commerce platform's monetization, it is important to predict high-level similarity more accurately than low-level similarity, as highly similar queries retrieves items according to user-intents, whereas moderately similar item retrieves related items, which may not lead to a purchase. Regression models fail to customize its loss func…
In this paper, we propose a new paradigm to learn audio features for Music Structure Analysis (MSA). We train a deep encoder to learn features such that the Self-Similarity-Matrix (SSM) resulting from those approximates a ground-truth SSM. This is done by minimizing a loss between both SSMs. Since this loss is differentiable w.r.t. its input features we can train the encoder in a straightforward way. We successfully demonstrate the use of this training paradigm using the Area Under the Curve ROC (AUC) on the RWC-Pop dataset.
ASUS Intelligent Cloud Services（华硕智能云服务）；Georgia Institute of Technology（佐治亚理工学院）
Single image dehazing is the ill-posed two-dimensional signal reconstruction problem. Recently, deep convolutional neural networks (CNN) have been successfully used in many computer vision problems. In this paper, we propose a Y-net that is named for its structure. This network reconstructs clear images by aggregating multi-scale features maps. Additionally, we propose a Wavelet Structure SIMilarity (W-SSIM) loss function in the training step. In the proposed loss function, discrete wavelet transforms are applied repeatedly to divide the image into differently sized patches with different freq…
University of Johannesburg（约翰内斯堡大学）；Institute for Intelligent Systems（智能系统研究所）
Sports data is more readily available and consequently, there has been an increase in the amount of sports analysis, predictions and rankings in the literature. Sports are unique in their respective stochastic nature, making analysis, and accurate predictions valuable to those involved in the sport. In response, we focus on Siamese Neural Networks (SNN) in unison with LightGBM and XGBoost models, to predict the importance of matches and to rank teams in Rugby and Basketball. Six models were developed and compared, a LightGBM, a XGBoost, a LightGBM (Contrastive Loss), LightGBM (Triplet Loss), a…
Universiti Malaya（马来亚大学）；University of Surrey（萨里大学）；Geek+（极智嘉）；iFlyTek-Surrey Joint Research Centre on Artificial Intelligence（科大讯飞-萨里人工智能联合研究中心）
A deep hashing model typically has two main learning objectives: to make the learned binary hash codes discriminative and to minimize a quantization error. With further constraints such as bit balance and code orthogonality, it is not uncommon for existing models to employ a large number (>4) of losses. This leads to difficulties in model training and subsequently impedes their effectiveness. In this work, we propose a novel deep hashing model with only a single learning objective. Specifically, we show that maximizing the cosine similarity between the continuous codes and their corresponding …
iCetana Pty Ltd（iCetana私人有限公司）；The University of Western Australia（西澳大利亚大学）
From video streaming to security and surveillance applications, video data play an important role in our daily living today. However, managing a large amount of video data and retrieving the most useful information for the user remain a challenging task. In this paper, we propose a novel video classification system that would benefit the scene understanding task. We define our classification problem as classifying background and foreground motions using the same feature representation for outdoor scenes. This means that the feature representation needs to be robust enough and adaptable to diff…
In this paper, we propose the coarse-to-fine optimization for the task of speech enhancement. Cosine similarity loss [1] has proven to be an effective metric to measure similarity of speech signals. However, due to the large variance of the enhanced speech with even the same cosine similarity loss in high dimensional space, a deep neural network learnt with this loss might not be able to predict enhanced speech with good quality. Our coarse-to-fine strategy optimizes the cosine similarity loss for different granularities so that more constraints are added to the prediction from high dimension …
Image-generating machine learning models are typically trained with loss functions based on distance in the image space. This often leads to over-smoothed results. We propose a class of loss functions, which we call deep perceptual similarity metrics (DeePSiM), that mitigate this problem. Instead of computing distances in the image space, we compute distances between image features extracted by deep neural networks. This metric better reflects perceptually similarity of images and thus leads to better results. We show three applications: autoencoder training, a modification of a variational au…
Image Sciences Institute, University Medical Center Utrecht（乌得勒支大学医学中心影像科学研究所）；Department of Radiology, University Medical Center Utrecht（乌得勒支大学医学中心放射科）；Department of Neurology and Neurosurgery, UMC Utrecht Brain Center, University Medical Center Utrecht（乌得勒支大学医学中心神经病学与神经外科系，乌得勒支大学医学中心脑中心）
Time-of-Flight Magnetic Resonance Angiographs (TOF-MRAs) enable visualization and analysis of cerebral arteries. This analysis may indicate normal variation of the configuration of the cerebrovascular system or vessel abnormalities, such as aneurysms. A model would be useful to represent normal cerebrovascular structure and variabilities in a healthy population and to differentiate from abnormalities. Current anomaly detection using autoencoding convolutional neural networks usually use a voxelwise mean-error for optimization. We propose optimizing a variational-autoencoder (VAE) with structur…
Meta-learning aims at optimizing the hyperparameters of a model class or training algorithm from the observation of data from a number of related tasks. Following the setting of Baxter [1], the tasks are assumed to belong to the same task environment, which is defined by a distribution over the space of tasks and by per-task data distributions. The statistical properties of the task environment thus dictate the similarity of the tasks. The goal of the meta-learner is to ensure that the hyperparameters obtain a small loss when applied for training of a new task sampled from the task environment…
Learning to hash is an efficient paradigm for exact and approximate nearest neighbor search from massive databases. Binary hash codes are typically extracted from an image by rounding output features from a CNN, which is trained on a supervised binary similar/ dissimilar task. Drawbacks of this approach are: (i) resulting codes do not necessarily capture semantic similarity of the input data (ii) rounding results in information loss, manifesting as decreased retrieval performance and (iii) Using only class-wise similarity as a target can lead to trivial solutions, simply encoding classifier ou…
Lakehead University（湖首大学）；University of Calgary（卡尔加里大学）
Objective: This paper presents an Alzheimer's disease (AD) detection method based on learning structural similarity between Magnetic Resonance Images (MRIs) and representing this similarity as a graph. Methods: We construct the similarity graph using embedded features of the input image (i.e., Non-Demented (ND), Very Mild Demented (VMD), Mild Demented (MD), and Moderated Demented (MDTD)). We experiment and compare different dimension-reduction and clustering algorithms to construct the best similarity graph to capture the similarity between the same class images using the cosine distance as a …
We introduce YOLO-pose, a novel heatmap-free approach for joint detection, and 2D multi-person pose estimation in an image based on the popular YOLO object detection framework. Existing heatmap based two-stage approaches are sub-optimal as they are not end-to-end trainable and training relies on a surrogate L1 loss that is not equivalent to maximizing the evaluation metric, i.e. Object Keypoint Similarity (OKS). Our framework allows us to train the model end-to-end and optimize the OKS metric itself. The proposed model learns to jointly detect bounding boxes for multiple persons and their corr…
Shanghai Institute for Advanced Communication and Data Science（上海先进通信与数据科学研究院）；Shanghai University（上海大学）
Speaker embeddings become growing popular in the text-independent speaker verification task. In this paper, we propose two improvements during the training stage. The improvements are both based on triplet cause the training stage and the evaluation stage of the baseline x-vector system focus on different aims. Firstly, we introduce triplet loss for optimizing the Euclidean distances between embeddings while minimizing the multi-class cross entropy loss. Secondly, we design an embedding similarity measurement network for controlling the similarity between the two selected embeddings. We furthe…
