Well-designed and trained deep neural networks can solve inverse electromagnetic problems much faster than conventional solvers. However, they need a physics framework to ensure producing physically correct results. Since most physics-guided deep learning inverse solvers require substantial training with numerous epochs, each involving solving a forward problem, their accuracy and efficiency are largely defined by the utilized forward solver, which becomes a bottleneck for their practical training. Thus, a fast and accurate self-supervised deep learning forward solver is presented. The solver uses a physics-based framework that divides the domain into two regions: an interior region, which includes any scatterers, and an exterior region, which represents the background medium. A hybrid loss function, incorporating Maxwell's curl equation and integral equation with the well-defined scalar background's Green's function, is employed to guide the scattered field generated from the neural network, ensuring global and local accuracy. To verify the generality of the solver, it is trained on random objects and tested on realistic models, showing high global and local metrics accuracy. For example, more than 95% of testing cases using the proposed method achieve less than 0.15 root-mean-square error in the calculated scattered field and dielectric properties of the imaged domain compared to the ground truth. In contrast, two recent deep learning methods could only realize that level of accuracy for less than 50% of the tested cases. The reported method is 97% faster than conventional solvers, enabling the development of reliable deep-learning inverse solvers.

Deep learning has the potential to address the bottleneck of conventional medical microwave tomography, which is ill-posed and has a high computation cost. However, current physics-guided deep learning methods may fail to capture the imaged object's salient regions, resulting in misdiagnosis. A deep neural network inspired by the distorted Born iterative method (DBIM) is proposed to address this challenge. This method, which avoids using Green's function, provides a theoretical explanation of why current deep learning methods guided by iterative physics algorithms fail in detecting abnormal tissues, making them unsuitable for real-life clinical applications. The proposed method consists of two main components: a set of forward neural network solvers and a series of inverse neural network blocks for updating the dielectric contrast of the imaging domain. The training of the network, designed to emulate the DBIM framework, is regularized by a hybrid loss function composed of two supervised and one self-supervised function. Each iteration in DBIM is performed using a neural network block with parameters different from those of other blocks, forming a sequential iterative approach. By calculating the perturbations in the electrical properties' profiles at each iteration, the proposed network can accurately reconstruct abnormal tissues associated with signals masked by those from healthy tissues. Assessments of the proposed method using the relative error, structure similarity index measure, Dice similarity coefficient, and Hausdorff distance show significant enhancements (19%, 18%, 40%, and 72%, respectively) compared to two recent deep learning-based microwave medical imaging algorithms.

Machine unlearning aims to remove knowledge of the specific training data in a well-trained model. Currently, machine unlearning methods typically handle all forgetting data in a single batch, removing the corresponding knowledge all at once upon request. However, in practical scenarios, unlearning requests often arise in a stream rather than in a single batch, leading to reduced efficiency and effectiveness of existing methods. Such challenges of streaming forgetting have not been the focus of much research. In this paper, to address the challenges of performance maintenance, efficiency, and data access in streaming unlearning, we introduce a streaming unlearning paradigm, formalizing the unlearning as a distribution shift problem. We then estimate the altered distribution and propose a novel streaming unlearning algorithm to achieve efficient streaming forgetting without requiring access to the original training data. Theoretical analyses confirm an O(√T + VT) error bound on the streaming unlearning regret, where VT represents the cumulative total variation in the optimal solution over T learning rounds. This theoretical guarantee is achieved under mild conditions without the strong restriction of convex loss function. Experiments across various models and datasets validate the performance of our proposed method.

Machine unlearning aims to remove information derived from forgotten data while preserving that of the remaining dataset in a well-trained model. With the increasing emphasis on data privacy, several approaches to machine unlearning have emerged. However, these methods typically rely on complete supervision throughout the unlearning process. Unfortunately, obtaining such supervision, whether for the forgetting or remaining data, can be impractical due to the substantial cost associated with annotating real-world datasets. This challenge prompts us to propose a supervision-free unlearning approach that operates without the need for labels during the unlearning process. Specifically, we introduce a variational approach to approximate the distribution of representations for the remaining data. Leveraging this approximation, we adapt the original model to eliminate information from the forgotten data at the representation level. To further address the issue of lacking supervision information, which hinders alignment with ground truth, we introduce a contrastive loss to facilitate the matching of representations between the remaining data and those of the original model, thus preserving predictive performance. Experimental results across various unlearning tasks demonstrate the effectiveness of our proposed method, Label-Agnostic Forgetting (LAF) without using any labels, which achieves comparable performance to state-of-the-art methods that rely on full supervision information. Furthermore, our approach excels in semi-supervised scenarios, leveraging limited supervision information to outperform fully supervised baselines. This work not only showcases the viability of supervision-free unlearning in deep models but also opens up a new possibility for future research in unlearning at the representation level.

Machine unlearning requires removing the information of forgetting data while keeping the necessary information of remaining data. Despite recent advancements in this area, existing methodologies mainly focus on the effect removal of forgetting data without considering the negative impact this can have on the information of the remaining data, resulting in significant performance degradation after data removal. Although some methods try to repair the performance of remaining data after removal, the forgotten information can also return after repair. Such an issue is due to the intricate intertwining of the forgetting and remaining data. Without adequately differentiating the influence of these two kinds of data on the model, existing algorithms take the risk of either inadequate removal of the forgetting data or unnecessary loss of valuable information from the remaining data. To address this shortcoming, the present study undertakes a causal analysis of the unlearning and introduces a novel framework termed Causal Machine Unlearning (CaMU). This framework adds intervention on the information of remaining data to disentangle the causal effects between forgetting data and remaining data. Then CaMU eliminates the causal impact associated with forgetting data while concurrently preserving the causal relevance of the remaining data. Comprehensive empirical results on various datasets and models suggest that CaMU enhances performance on the remaining data and effectively minimizes the influences of forgetting data. Notably, this work is the first to interpret deep model unlearning tasks from a new perspective of causality and provide a solution based on causal analysis, which opens up new possibilities for future research in deep model unlearning.

When deploying deep neural networks, quantification of a model's uncertainty is necessary to provide confidence in its predictions by distinguishing between accurate predictions and coincidentally correct guesses. While it is known that the accuracy of predictions is dependent on the data on which the model was trained, to date, limited work has examined the relationship between training data quantity and uncertainty quantification. In this paper, we propose two metrics to assess the ‘quality’ of uncertainty quantification, and investigate the relationship between training data quantity and Monte Carlo Dropout uncertainty quantification in supervised and semi-supervised learning across various text-based datasets. We found that in supervised learning, uncertainty quan-tification quality (across both metrics) initially increased for larger quantities of training data, but interestingly, after a certain threshold, began to gradually decline. In semi-supervised learning, uncertainty quantification was enhanced by both a greater number of training samples and greater proportion of pre-labelled data. These results suggest that for supervised learning, data scientists generally ought not to invest resources into acquiring more training data solely for superior uncertainty quantification. However, if semi-supervised learning is necessary, then there is a marked benefit in obtaining more data.

Abstract Text classification is a widely used task in natural language processing. However, the presence of stereotype bias in text classification can lead to unfair and inaccurate predictions. Stereotype bias is particularly prevalent in words that are unevenly distributed across classes and are associated with specific categories. This bias can be further strengthened in pre-trained models on large natural language datasets. Prior works to remove stereotype bias have mainly focused on specific demographic groups or relied on specific thesauri without measuring the influence of stereotype words on predictions. In this work, we present a causal analysis of how stereotype bias occurs and affects text classification, and propose a framework to mitigate stereotype bias. Our framework detects potential stereotype bias words using SHAP values and alleviates bias in the prediction stage through a counterfactual approach. Unlike existing debiasing methods, our framework does not rely on existing stereotype word sets and can dynamically evaluate the influence of words on stereotype bias. Extensive experiments and ablation studies show that our approach effectively improves classification performance while mitigating stereotype bias.

To drive safely requires perceiving vast amounts of rapidly changing visual information. This can exhaust our limited perceptual capacity and lead to cases of 'looking but failing to see', reportedly the third largest contributing factor to road traffic accidents. In the present work we use a 3D convolutional neural network to model the perceptual demand of varied driving situations. To validate the method we introduce a new labelled dataset of approximately 2300 videos of driving in Brussels and California.

In team sports like soccer, utilizing tracking data for analysis is challenging due to the dynamic and multi-agent nature of the data. The biggest issue surrounds the changing of positions or “roles” between players on a frame-to-frame basis, which causes misalignment of the data and makes it difficult to perform team analysis. In this paper, we present an unsupervised method to learn a formation template which allows us to “align” the tracking data at the frame level. Not only does this approach give important contextual information to facilitate large-scale analysis (e.g., we know when a player is in the left-wing position compared to left-back), it also yields the team structure or “formation” which serves as a strong descriptor for identifying a team's style. The utility of the approach is demonstrated on a full season of player and ball tracking data from a professional soccer league consisting of over 21.5 million frames of player tracking data.

Although the collection of player and ball tracking data is fast becoming the norm in professional sports, large-scale mining of such spatiotemporal data has yet to surface. In this paper, given an entire season's worth of player and ball tracking data from a professional soccer league (≈400,000,000 data points), we present a method which can conduct both individual player and team analysis. Due to the dynamic, continuous and multi-player nature of team sports like soccer, a major issue is aligning player positions over time. We present a "role-based" representation that dynamically updates each player's relative role at each frame and demonstrate how this captures the short-term context to enable both individual player and team analysis. We discover role directly from data by utilizing a minimum entropy data partitioning method and show how this can be used to accurately detect and visualize formations, as well as analyze individual player behavior.

We consider the problem of learning predictive models for in-game sports play prediction. Focusing on basketball, we develop models for anticipating near-future events given the current game state. We employ a latent factor modeling approach, which leads to a compact data representation that enables efficient prediction given raw spatiotemporal tracking data. We validate our approach using tracking data from the 2012-2013 NBA season, and show that our model can make accurate in-game predictions. We provide a detailed inspection of our learned factors, and show that our model is interpretable and corresponds to known intuitions of basketball game play.

In this paper, we describe a method to represent and discover adversarial group behavior in a continuous domain. In comparison to other types of behavior, adversarial behavior is heavily structured as the location of a player (or agent) is dependent both on their teammates and adversaries, in addition to the tactics or strategies of the team. We present a method which can exploit this relationship through the use of a spatiotemporal basis model. As players constantly change roles during a match, we show that employing a "role-based" representation instead of one based on player "identity" can best exploit the playing structure. As vision-based systems currently do not provide perfect detection/tracking (e.g. missed or false detections), we show that our compact representation can effectively "denoise" erroneous detections as well as enabling temporal analysis, which was previously prohibitive due to the dimensionality of the signal. To evaluate our approach, we used a fully instrumented field-hockey pitch with 8 fixed high-definition (HD) cameras and evaluated our approach on approximately 200,000 frames of data from a state-of-the-art real-time player detector and compare it to manually labelled data.

Real-world AI systems have been recently deployed which can automatically analyze the plan and tactics of tennis players. As the game-state is updated regularly at short intervals (i.e. point-level), a library of successful and unsuccessful plans of a player can be learnt over time. Given the relative strengths and weaknesses of a player’s plans, a set of proven plans or tactics from the library that characterize a player can be identified. For low-scoring, continuous team sports like soccer, such analysis for multi-agent teams does not exist as the game is not segmented into “discretized” plays (i.e. plans), making it difficult to obtain a library that characterizes a team’s behavior. Additionally, as player tracking data is costly and difficult to obtain, we only have partial team tracings in the form of ball actions which makes this problem even more difficult. In this paper, we propose a method to overcome these issues by representing team behavior via play-segments, which are spatio-temporal descriptions of ball movement over fixed windows of time. Using these representations we can characterize team behavior from entropy maps, which give a measure of predictability of team behaviors across the field. We show the efficacy and applicability of our method on the 2010-2011 English Premier League soccer data.

CCTV and surveillance networks are increasingly being used for operational as well as security tasks. One emerging area of technology that lends itself to operational analytics is soft biometrics. Soft biometrics can be used to describe a person and detect them throughout a sparse multi-camera network. This enables them to be used to perform tasks such as determining the time taken to get from point to point, and the paths taken through an environment by detecting and matching people across disjoint views. However, in a busy environment where there are 100's if not 1000's of people such as an airport, attempting to monitor everyone is highly unrealistic. In this paper we propose an average soft biometric, that can be used to identity people who look distinct, and are thus suitable for monitoring through a large, sparse camera network. We demonstrate how an average soft biometric can be used to identify unique people to calculate operational measures such as the time taken to travel from point to point.