Modern image retrieval systems increasingly rely on the use of deep neural networks to learn embedding spaces in which distance encodes the relevance between a given query and image. In this setting, existing approaches tend to emphasize one of two properties. Triplet-based methods capture top-$k$ relevancy, where all top-$k$ scoring documents are assumed to be relevant to a given query Pairwise contrastive models capture threshold relevancy, where all documents scoring higher than some threshold are assumed to be relevant. In this paper, we propose Cross-Example Softmax which combines the properties of top-$k$ and threshold relevancy. In each iteration, the proposed loss encourages all queries to be closer to their matching images than all queries are to all non-matching images. This leads to a globally more calibrated similarity metric and makes distance more interpretable as an absolute measure of relevance. We further introduce Cross-Example Negative Mining, in which each pair is compared to the hardest negative comparisons across the entire batch. Empirically, we show in a series of experiments on Conceptual Captions and Flickr30k, that the proposed method effectively improves global calibration and also retrieval performance.
Peking University（北京大学）；University of California, Los Angeles（加州大学洛杉矶分校）
Contextual representation models have achieved great success in improving various downstream tasks. However, these language-model-based encoders are difficult to train due to the large parameter sizes and high computational complexity. By carefully examining the training procedure, we find that the softmax layer (the output layer) causes significant inefficiency due to the large vocabulary size. Therefore, we redesign the learning objective and propose an efficient framework for training contextual representation models. Specifically, the proposed approach bypasses the softmax layer by perform…
In-context classification: Calibration and noise in linear attention with softmax readout
Harvard University（哈佛大学）；Williams College（威廉姆斯学院）；Kempner Institute, Harvard University（哈佛肯珀研究所）；Center for Brain Science and Society of Fellows, Harvard University（哈佛大学生脑科学与社会研究员院）
Microsoft Research Asia（微软亚洲研究院）；MILA（米拉研究所）；S-Lab, Nanyang Technological University（南洋理工大学S-Lab实验室）；Nanyang Technological University（南洋理工大学）
Confidence calibration is of great importance to the reliability of decisions made by machine learning systems. However, discriminative classifiers based on deep neural networks are often criticized for producing overconfident predictions that fail to reflect the true correctness likelihood of classification accuracy. We argue that such an inability to model uncertainty is mainly caused by the closed-world nature in softmax: a model trained by the cross-entropy loss will be forced to classify input into one of $K$ pre-defined categories with high probability. To address this problem, we for th…
University of Minnesota（明尼苏达大学）；National University of Singapore（新加坡国立大学）
Intra-class compactness and inter-class separability are crucial indicators to measure the effectiveness of a model to produce discriminative features, where intra-class compactness indicates how close the features with the same label are to each other and inter-class separability indicates how far away the features with different labels are. In this work, we investigate intra-class compactness and inter-class separability of features learned by convolutional networks and propose a Gaussian-based softmax ($\mathcal{G}$-softmax) function that can effectively improve intra-class compactness and …
University of Amsterdam（阿姆斯特丹大学）；Universidad Carlos III de Madrid（马德里卡洛斯三世大学）
We study the statistical properties of learning to defer (L2D) to multiple experts. In particular, we address the open problems of deriving a consistent surrogate loss, confidence calibration, and principled ensembling of experts. Firstly, we derive two consistent surrogates -- one based on a softmax parameterization, the other on a one-vs-all (OvA) parameterization -- that are analogous to the single expert losses proposed by Mozannar and Sontag (2020) and Verma and Nalisnick (2022), respectively. We then study the frameworks' ability to estimate P( m_j = y | x ), the probability that the jth…
Dirichlet-Based Prediction Calibration for Learning with Noisy Labels
College of Computer Science and Technology/Artificial Intelligence, Nanjing University of Aeronautics and Astronautics（南京航空航天大学计算机科学与技术学院/人工智能学院）；MIIT Key Laboratory of Pattern Analysis and Machine Intelligence（工业和信息化部模式分析与机器智能重点实验室）
PSL: Rethinking and Improving Softmax Loss from Pairwise Perspective for Recommendation
State Key Laboratory of Blockchain and Data Security, Zhejiang University（区块链与数据安全国家重点实验室，浙江大学）；College of Computer Science and Technology, Zhejiang University（计算机科学与技术学院，浙江大学）
The great performances of deep learning are undeniable, with impressive results over a wide range of tasks. However, the output confidence of these models is usually not well-calibrated, which can be an issue for applications where confidence on the decisions is central to providing trust and reliability (e.g., autonomous driving or medical diagnosis). For models using softmax at the last layer, Temperature Scaling (TS) is a state-of-the-art calibration method, with low time and memory complexity as well as demonstrated effectiveness. TS relies on a T parameter to rescale and calibrate values …
Google Research（谷歌研究院）；Google DeepMind（谷歌DeepMind）
Dual encoder models are ubiquitous in modern classification and retrieval. Crucial for training such dual encoders is an accurate estimation of gradients from the partition function of the softmax over the large output space; this requires finding negative targets that contribute most significantly ("hard negatives"). Since dual encoder model parameters change during training, the use of traditional static nearest neighbor indexes can be sub-optimal. These static indexes (1) periodically require expensive re-building of the index, which in turn requires (2) expensive re-encoding of all targets…
Obtaining reliable and accurate quantification of uncertainty estimates from deep neural networks is important in safety-critical applications. A well-calibrated model should be accurate when it is certain about its prediction and indicate high uncertainty when it is likely to be inaccurate. Uncertainty calibration is a challenging problem as there is no ground truth available for uncertainty estimates. We propose an optimization method that leverages the relationship between accuracy and uncertainty as an anchor for uncertainty calibration. We introduce a differentiable accuracy versus uncert…
Neural networks utilize the softmax as a building block in classification tasks, which contains an overconfidence problem and lacks an uncertainty representation ability. As a Bayesian alternative to the softmax, we consider a random variable of a categorical probability over class labels. In this framework, the prior distribution explicitly models the presumed noise inherent in the observed label, which provides consistent gains in generalization performance in multiple challenging tasks. The proposed method inherits advantages of Bayesian approaches that achieve better uncertainty estimation…
PRECISE Center（PRECISE中心）；University of Pennsylvania（宾夕法尼亚大学）
Providing reliable model uncertainty estimates is imperative to enabling robust decision making by autonomous agents and humans alike. While recently there have been significant advances in confidence calibration for trained models, examples with poor calibration persist in most calibrated models. Consequently, multiple techniques have been proposed that leverage label-invariant transformations of the input (i.e., an input manifold) to improve worst-case confidence calibration. However, manifold-based confidence calibration techniques generally do not scale and/or require expensive retraining …
Traditional federated optimization methods perform poorly with heterogeneous data (ie, accuracy reduction), especially for highly skewed data. In this paper, we investigate the label distribution skew in FL, where the distribution of labels varies across clients. First, we investigate the label distribution skew from a statistical view. We demonstrate both theoretically and empirically that previous methods based on softmax cross-entropy are not suitable, which can result in local models heavily overfitting to minority classes and missing classes. Additionally, we theoretically introduce a dev…
University of Amsterdam（阿姆斯特丹大学）；Inception Institute of Artificial Intelligence（起源人工智能研究院）
Multi-task learning aims to explore task relatedness to improve individual tasks, which is of particular significance in the challenging scenario that only limited data is available for each task. To tackle this challenge, we propose variational multi-task learning (VMTL), a general probabilistic inference framework for learning multiple related tasks. We cast multi-task learning as a variational Bayesian inference problem, in which task relatedness is explored in a unified manner by specifying priors. To incorporate shared knowledge into each task, we design the prior of a task to be a learna…
Unpacking Softmax: How Temperature Drives Representation Collapse, Compression, and Generalization
Warsaw University of Technology（华沙技术大学）；University of Basel（巴塞尔大学）；Mila, Québec AI Institute（魁北克人工智能研究所）；IDEAS Research Institute（IDEAS研究所）；Tooploox（Tooploox公司）
A wide breadth of research has devised data augmentation approaches that can improve both accuracy and generalization performance for neural networks. However, augmented data can end up being far from the clean training data and what is the appropriate label is less clear. Despite this, most existing work simply uses one-hot labels for augmented data. In this paper, we show re-using one-hot labels for highly distorted data might run the risk of adding noise and degrading accuracy and calibration. To mitigate this, we propose a generic method AutoLabel to automatically learn the confidence in t…
National Taiwan University（台湾大学）；Carnegie Mellon University（卡内基梅隆大学）
Generative Adversarial Networks (GANs) is a powerful family of models that learn an underlying distribution to generate synthetic data. Many existing studies of GANs focus on improving the realness of the generated image data for visual applications, and few of them concern about improving the quality of the generated data for training other classifiers -- a task known as the model compatibility problem. As a consequence, existing GANs often prefer generating `easier' synthetic data that are far from the boundaries of the classifiers, and refrain from generating near-boundary data, which are k…
Recently, deep neural networks have become to be used in a variety of applications. While the accuracy of deep neural networks is increasing, the confidence score, which indicates the reliability of the prediction results, is becoming more important. Deep neural networks are seen as highly accurate but known to be overconfident, making it important to calibrate the confidence score. Many studies have been conducted on confidence calibration. They calibrate the confidence score of the model to match its accuracy, but it is not clear whether these confidence scores can improve the performance of…
Datasets with noisy labels are a common occurrence in practical applications of classification methods. We propose a simple probabilistic method for training deep classifiers under input-dependent (heteroscedastic) label noise. We assume an underlying heteroscedastic generative process for noisy labels. To make gradient based training feasible we use a temperature parameterized softmax as a smooth approximation to the assumed generative process. We illustrate that the softmax temperature controls a bias-variance trade-off for the approximation. By tuning the softmax temperature, we improve acc…
Let the Experts Speak: Improving Survival Prediction & Calibration via Mixture-of-Experts Heads
Department of Computer Science Columbia University USA（哥伦比亚大学计算机科学系）；Department of Computer Science New York University USA（纽约大学计算机科学系）；Department of Neurology Columbia University Medical Center USA（哥伦比亚大学医学中心神经病学系）；Department of Computer Science Barnard College USA（巴纳德学院计算机科学系）；Department of Biomedical Informatics Columbia University Medical Center USA（哥伦比亚大学医学中心生物医学信息学系）；NewYork-Presbyterian Hospital at Columbia University Medical Center USA（哥伦比亚大学医学中心新英格兰-纽约 Presbyterian 医院）
Neural networks have proven successful at learning from complex data distributions by acting as universal function approximators. However, they are often overconfident in their predictions, which leads to inaccurate and miscalibrated probabilistic predictions. The problem of overconfidence becomes especially apparent in cases where the test-time data distribution differs from that which was seen during training. We propose a solution to this problem by seeking out regions of feature space where the model is unjustifiably overconfident, and conditionally raising the entropy of those predictions…
Calibration and uncertainty estimation are crucial topics in high-risk environments. We introduce a new diversity regularizer for classification tasks that uses out-of-distribution samples and increases the overall accuracy, calibration and out-of-distribution detection capabilities of ensembles. Following the recent interest in the diversity of ensembles, we systematically evaluate the viability of explicitly regularizing ensemble diversity to improve calibration on in-distribution data as well as under dataset shift. We demonstrate that diversity regularization is highly beneficial in archit…
School of Computer Science, Machine Learning Department, Carnegie Mellon University（卡内基梅隆大学计算机科学学院机器学习系）
Deep Neural Networks (DNNs) often struggle with one-shot learning where we have only one or a few labeled training examples per category. In this paper, we argue that by using side information, we may compensate the missing information across classes. We introduce two statistical approaches for fusing side information into data representation learning to improve one-shot learning. First, we propose to enforce the statistical dependency between data representations and multiple types of side information. Second, we introduce an attention mechanism to efficiently treat examples belonging to the …
Arizona State University（亚利桑那州立大学）；Lawrence Livermore National Labs（劳伦斯利弗莫尔国家实验室）；IBM Research AI（IBM研究院人工智能部门）
The hypothesis that sub-network initializations (lottery) exist within the initializations of over-parameterized networks, which when trained in isolation produce highly generalizable models, has led to crucial insights into network initialization and has enabled efficient inferencing. Supervised models with uncalibrated confidences tend to be overconfident even when making wrong prediction. In this paper, for the first time, we study how explicit confidence calibration in the over-parameterized network impacts the quality of the resulting lottery tickets. More specifically, we incorporate a s…
University at Buffalo（布法罗大学）；Penn State Erie（宾夕法尼亚州立大学伊利分校）
Deep Neural Networks (DNNs), despite their tremendous success in recent years, could still cast doubts on their predictions due to the intrinsic uncertainty associated with their learning process. Ensemble techniques and post-hoc calibrations are two types of approaches that have individually shown promise in improving the uncertainty calibration of DNNs. However, the synergistic effect of the two types of methods has not been well explored. In this paper, we propose a truth discovery framework to integrate ensemble-based and post-hoc calibration methods. Using the geometric variance of the en…
On Impact of Loss Function on the Performance of Neural Networks in Melanoma Diagnosis
Technological University Dublin（都柏林理工大学）；Research Ireland Centre for Research Training in Machine Learning, ML-Labs（爱尔兰研究中心机器学习研究培训中心，机器学习实验室）；Dublin City University（都柏林城市大学）；University College Dublin（都柏林大学学院）
University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
Deep neural network (DNN) classifiers are often overconfident, producing miscalibrated class probabilities. In high-risk applications like healthcare, practitioners require $\textit{fully calibrated}$ probability predictions for decision-making. That is, conditioned on the prediction $\textit{vector}$, $\textit{every}$ class' probability should be close to the predicted value. Most existing calibration methods either lack theoretical guarantees for producing calibrated outputs, reduce classification accuracy in the process, or only calibrate the predicted class. This paper proposes a new Kerne…
Neural networks lack adversarial robustness, i.e., they are vulnerable to adversarial examples that through small perturbations to inputs cause incorrect predictions. Further, trust is undermined when models give miscalibrated predictions, i.e., the predicted probability is not a good indicator of how much we should trust our model. In this paper, we study the connection between adversarial robustness and calibration and find that the inputs for which the model is sensitive to small perturbations (are easily attacked) are more likely to have poorly calibrated predictions. Based on this insight…
Centro de Informática, Universidade Federal de Pernambuco（伯南布哥联邦大学信息中心）；Montreal Institute for Learning Algorithms (MILA), Université de Montréal (UdeM)（蒙特利尔大学蒙特利尔学习算法研究所（MILA））
In this paper, we argue that the unsatisfactory out-of-distribution (OOD) detection performance of neural networks is mainly due to the SoftMax loss anisotropy and propensity to produce low entropy probability distributions in disagreement with the principle of maximum entropy. Current out-of-distribution (OOD) detection approaches usually do not directly fix the SoftMax loss drawbacks, but rather build techniques to circumvent it. Unfortunately, those methods usually produce undesired side effects (e.g., classification accuracy drop, additional hyperparameters, slower inferences, and collecti…
Habana Labs – An Intel company（Habana Labs——英特尔旗下公司）；Technion（以色列理工学院）
Lately, post-training quantization methods have gained considerable attention, as they are simple to use, and require only a small unlabeled calibration set. This small dataset cannot be used to fine-tune the model without significant over-fitting. Instead, these methods only use the calibration set to set the activations' dynamic ranges. However, such methods always resulted in significant accuracy degradation, when used below 8-bits (except on small datasets). Here we aim to break the 8-bit barrier. To this end, we minimize the quantization errors of each layer separately by optimizing its p…
Improving Perturbation-based Explanations by Understanding the Role of Uncertainty Calibration
Siemens AG（西门子股份公司）；LMU Munich（慕尼黑大学）；Munich Center for Machine Learning (MCML)（慕尼黑机器学习中心）；Goethe University Frankfurt（法兰克福歌德大学）；German Cancer Research Center (DKFZ)（德国癌症研究中心）；German Cancer Consortium (DKTK)（德国癌症联合会）
Amazon Research（亚马逊研究院）；Amazon Web Services（亚马逊云科技）
The Gaussian mechanism is an essential building block used in multitude of differentially private data analysis algorithms. In this paper we revisit the Gaussian mechanism and show that the original analysis has several important limitations. Our analysis reveals that the variance formula for the original mechanism is far from tight in the high privacy regime ($\varepsilon \to 0$) and it cannot be extended to the low privacy regime ($\varepsilon \to \infty$). We address these limitations by developing an optimal Gaussian mechanism whose variance is calibrated directly using the Gaussian cumula…
Maximum Entropy on Erroneous Predictions (MEEP): Improving model calibration for medical image segmentation
Tryolabs（Tryolabs公司）；Research institute for signals, systems and computational intelligence, sinc(i), FICH-UNL / CONICET（信号、系统与计算智能研究所（sinc(i)，国立拉普拉塔大学计算机科学学院/阿根廷国家科学与技术研究理事会））；LIVIA, ETS Montreal（蒙特利尔高等技术学院LIVIA实验室）；International Laboratory on Learning Systems (ILLS)（国际学习系统实验室（ILLS））
