Recent work has shown that the performance of machine learning models can vary substantially when models are evaluated on data drawn from a distribution that is close to but different from the training distribution. As a result, predicting model performance on unseen distributions is an important challenge. Our work connects techniques from domain adaptation and predictive uncertainty literature, and allows us to predict model accuracy on challenging unseen distributions without access to labeled data. In the context of distribution shift, distributional distances are often used to adapt models and improve their performance on new domains, however accuracy estimation, or other forms of predictive uncertainty, are often neglected in these investigations. Through investigating a wide range of established distributional distances, such as Frechet distance or Maximum Mean Discrepancy, we determine that they fail to induce reliable estimates of performance under distribution shift. On the other hand, we find that the difference of confidences (DoC) of a classifier's predictions successfully estimates the classifier's performance change over a variety of shifts. We specifically investigate the distinction between synthetic and natural distribution shifts and observe that despite its simplicity DoC consistently outperforms other quantifications of distributional difference. $DoC$ reduces predictive error by almost half ($46\%$) on several realistic and challenging distribution shifts, e.g., on the ImageNet-Vid-Robust and ImageNet-Rendition datasets.
Max Planck Institute for Informatics（马克斯·普朗克信息学研究所）；University of Tübingen（图宾根大学）
Adversarial training yields robust models against a specific threat model, e.g., $L_\infty$ adversarial examples. Typically robustness does not generalize to previously unseen threat models, e.g., other $L_p$ norms, or larger perturbations. Our confidence-calibrated adversarial training (CCAT) tackles this problem by biasing the model towards low confidence predictions on adversarial examples. By allowing to reject examples with low confidence, robustness generalizes beyond the threat model employed during training. CCAT, trained only on $L_\infty$ adversarial examples, increases robustness ag…
Out-of-distribution (OOD) data poses serious challenges in deployed machine learning models as even subtle changes could incur significant performance drops. Being able to estimate a model's performance on test data is important in practice as it indicates when to trust to model's decisions. We present a simple yet effective method to predict a model's performance on an unknown distribution without any addition annotation. Our approach is rooted in the Optimal Transport theory, viewing test samples' output softmax scores from deep neural networks as empirical samples from an unknown distributi…
In this paper, we propose a confidence-calibration method for predicting the winner of a famous multiplayer online battle arena (MOBA) game, League of Legends. In MOBA games, the dataset may contain a large amount of input-dependent noise; not all of such noise is observable. Hence, it is desirable to attempt a confidence-calibrated prediction. Unfortunately, most existing confidence calibration methods are pertaining to image and document classification tasks where consideration on uncertainty is not crucial. In this paper, we propose a novel calibration method that takes data uncertainty int…
Tsinghua University（清华大学）；University of Cambridge（剑桥大学）
It is often remarked that neural networks fail to increase their uncertainty when predicting on data far from the training distribution. Yet naively using softmax confidence as a proxy for uncertainty achieves modest success in tasks exclusively testing for this, e.g., out-of-distribution (OOD) detection. This paper investigates this contradiction, identifying two implicit biases that do encourage softmax confidence to correlate with epistemic uncertainty: 1) Approximately optimal decision boundary structure, and 2) Filtering effects of deep networks. It describes why low-dimensional intuition…
To increase the trustworthiness of deep neural network (DNN) classifiers, an accurate prediction confidence that represents the true likelihood of correctness is crucial. Towards this end, many post-hoc calibration methods have been proposed to leverage a lightweight model to map the target DNN's output layer into a calibrated confidence. Nonetheless, on an out-of-distribution (OOD) dataset in practice, the target DNN can often mis-classify samples with a high confidence, creating significant challenges for the existing calibration methods to produce an accurate confidence. In this paper, we p…
In this work, a machine learning approach is developed for predicting the outcomes of football matches. The novelty of this research lies in the utilisation of the Kelly Index to first classify matches into categories where each one denotes the different levels of predictive difficulty. Classification models using a wide suite of algorithms were developed for each category of matches in order to determine the efficacy of the approach. In conjunction to this, a set of previously unexplored features were engineering including Elo-based variables. The dataset originated from the Premier League ma…
Real-world machine learning deployments are characterized by mismatches between the source (training) and target (test) distributions that may cause performance drops. In this work, we investigate methods for predicting the target domain accuracy using only labeled source data and unlabeled target data. We propose Average Thresholded Confidence (ATC), a practical method that learns a threshold on the model's confidence, predicting accuracy as the fraction of unlabeled examples for which model confidence exceeds that threshold. ATC outperforms previous methods across several model architectures…
Machines that can predict the effect of physical interactions on the dynamics of previously unseen object instances are important for creating better robots and interactive virtual worlds. In this work, we focus on predicting the dynamics of 3D objects on a plane that have just been subjected to an impulsive force. In particular, we predict the changes in state - 3D position, rotation, velocities, and stability. Different from previous work, our approach can generalize dynamics predictions to object shapes and initial conditions that were unseen during training. Our method takes the 3D object'…
We study the distribution and uncertainty of nonconvex optimization for noisy tensor completion -- the problem of estimating a low-rank tensor given incomplete and corrupted observations of its entries. Focusing on a two-stage estimation algorithm proposed by Cai et al. (2019), we characterize the distribution of this nonconvex estimator down to fine scales. This distributional theory in turn allows one to construct valid and short confidence intervals for both the unseen tensor entries and the unknown tensor factors. The proposed inferential procedure enjoys several important features: (1) it…
Accurate traffic prediction is a key ingredient to enable traffic management like rerouting cars to reduce road congestion or regulating traffic via dynamic speed limits to maintain a steady flow. A way to represent traffic data is in the form of temporally changing heatmaps visualizing attributes of traffic, such as speed and volume. In recent works, U-Net models have shown SOTA performance on traffic forecasting from heatmaps. We propose to combine the U-Net architecture with graph layers which improves spatial generalization to unseen road networks compared to a Vanilla U-Net. In particular…
University of Guelph（圭尔夫大学）；Vector Institute for Artificial Intelligence（向量人工智能研究所）；Canadian Institute for Advanced Research（加拿大高级研究所）
It has been suggested that adversarial examples cause deep learning models to make incorrect predictions with high confidence. In this work, we take the opposite stance: an overly confident model is more likely to be vulnerable to adversarial examples. This work is one of the most proactive approaches taken to date, as we link robustness with non-calibrated model confidence on noisy images, providing a data-augmentation-free path forward. The adversarial examples phenomenon is most easily explained by the trend of increasing non-regularized model capacity, while the diversity and number of sam…
Institute of Information Systems（信息系统研究所）；University of Liechtenstein（列支敦士登大学）
The data distribution commonly evolves over time leading to problems such as concept drift that often decrease classifier performance. Current techniques are not adequate for this problem because they either require detailed knowledge of the transformation or are not suited for anticipating unseen domains but can only adapt to domains, where data samples are available. We seek to predict unseen data (and their labels) allowing us to tackle challenges s a non-constant data distribution in a proactive manner rather than detecting and reacting to already existing changes that might already have l…
Center for Data Anaytics and Biomedical Informatics, Temple University（天普大学数据分析与生物医学信息学中心）
For highly sensitive real-world predictive analytic applications such as healthcare and medicine, having good prediction accuracy alone is often not enough. These kinds of applications require a decision making process which uses uncertainty estimation as input whenever possible. Quality of uncertainty estimation is a subject of over or under confident prediction, which is often not addressed in many models. In this paper we show several extensions to the Gaussian Conditional Random Fields model, which aim to provide higher quality uncertainty estimation. These extensions are applied to the te…
Hertie Institute for AI in Brain Health, University of Tübingen（蒂宾根大学赫蒂脑健康人工智能研究所）；Tübingen AI Center, University of Tübingen（蒂宾根大学蒂宾根人工智能中心）；Charité – Universitätsmedizin Berlin, Department of Psychiatry and Psychotherapy（柏林夏里特医科大学精神病学与心理治疗系）；German Center for Mental Health (DZPG), partner site Tübingen（德国精神卫生中心（DZPG）蒂宾根合作站点）
Central Electronics Engineering Research Institute（中央电子工程研究所）；Academy of Scientific and Innovative Research（科学与创新研究院）
This letter proposes a novel deep learning framework (DLF) that addresses two major hurdles in the adoption of deep learning techniques for solving physics-based problems: 1) requirement of the large dataset for training the DL model, 2) consistency of the DL model with the physics of the phenomenon. The framework is generic in nature and can be applied to model a phenomenon from other fields of research too as long as its behaviour is known. To demonstrate the technique, a semi-supervised physics guided neural network (SPGNN) has been developed that predicts I-V characteristics of a gallium n…
MDQEC-QAS: Meta-Decoding for Quantum Error Correction with Hardware-Aware VQC Search and Confidence-Gated Recovery
Department of Physics, Indian Institute of Technology (BHU)（印度理工学院（BHU）物理系）；eBRAIN Lab, Division of Engineering, New York University Abu Dhabi (NYUAD)（纽约大学阿布扎克分校（NYUAD）工程学院eBRAIN实验室）；Center for Quantum and Topological Systems (CQTS), NYUAD Research Institute, NYUAD（纽约大学阿布扎克分校（NYUAD）量子与拓扑系统中心（CQTS））
The necessity of predictive models in the drug discovery industry cannot be understated. With the sheer volume of potentially useful compounds that are considered for use, it is becoming increasingly computationally difficult to investigate the overlapping interactions between drugs. Understanding this is also important to the layperson who needs to know what they can and cannot mix, especially for those who use recreational drugs - which do not have the same rigorous warnings as prescription drugs. Without access to deterministic, experimental results for every drug combination, other methods…
Cyclica（赛克利卡公司）；University of Toronto（多伦多大学）；Princess Margaret Cancer Centre（玛格丽特公主癌症中心）；University Health Network（大学健康网络）；Vector Institute（向量研究所）
Performance of neural network models relies on the availability of large datasets with minimal levels of uncertainty. Transfer Learning (TL) models have been proposed to resolve the issue of small dataset size by letting the model train on a bigger, task-related reference dataset and then fine-tune on a smaller, task-specific dataset. In this work, we apply a transfer learning approach to improve predictive power in noisy data systems with large variable confidence datasets. We propose a deep neural network method called Filtered Transfer Learning (FTL) that defines multiple tiers of data conf…
Recent advances in the e-commerce fashion industry have led to an exploration of novel ways to enhance buyer experience via improved personalization. Predicting a proper size for an item to recommend is an important personalization challenge, and is being studied in this work. Earlier works in this field either focused on modeling explicit buyer fitment feedback or modeling of only a single aspect of the problem (e.g., specific category, brand, etc.). More recent works proposed richer models, either content-based or sequence-based, better accounting for content-based aspects of the problem or …
Many state-of-the-art algorithms for solving hard combinatorial problems in artificial intelligence (AI) include elements of stochasticity that lead to high variations in runtime, even for a fixed problem instance. Knowledge about the resulting runtime distributions (RTDs) of algorithms on given problem instances can be exploited in various meta-algorithmic procedures, such as algorithm selection, portfolios, and randomized restarts. Previous work has shown that machine learning can be used to individually predict mean, median and variance of RTDs. To establish a new state-of-the-art in predic…
Background: Overweight and obesity are an increasing phenomenon worldwide. Predicting future overweight or obesity early in the childhood reliably could enable a successful intervention by experts. While a lot of research has been done using explanatory modeling methods, capability of machine learning, and predictive modeling, in particular, remain mainly unexplored. In predictive modeling models are validated with previously unseen examples, giving a more accurate estimate of their performance and generalization ability in real-life scenarios. Objective: To find and review existing overweight…
Shortest-Path Flow Matching with Mixture-Conditioned Bases for OOD Generalization to Unseen Conditions
Wellcome Sanger Institute（Wellcome桑格研究所）；Institute for Molecular Medicine Finland (FIMM)（芬兰分子医学研究所）；University of Helsinki（赫尔辛基大学）；School of Computing and Information Systems（计算与信息系统学院）；University of Cambridge（剑桥大学）；Cambridge Centre for AI in Medicine（剑桥人工智能医学中心）；MediaTek Research（MediaTek研究）；Cambridge Stem Cell Institute（剑桥干细胞研究所）
Stable-Shift: Biologically Structured Prediction of Transcriptional Responses to Unseen Gene Perturbations
Department of Computer Science, Virginia Tech（弗吉尼亚理工大学计算机科学系）；Fralin Biomedical Research Institute, Virginia Tech（弗吉尼亚理工大学弗拉林生物医学研究中心）；FBRI Cancer Research Center（FBRI癌症研究中心）
