The prevalence of employing attention mechanisms has brought along concerns on the interpretability of attention distributions. Although it provides insights about how a model is operating, utilizing attention as the explanation of model predictions is still highly dubious. The community is still seeking more interpretable strategies for better identifying local active regions that contribute the most to the final decision. To improve the interpretability of existing attention models, we propose a novel Bilinear Representative Non-Parametric Attention (BR-NPA) strategy that captures the task-relevant human-interpretable information. The target model is first distilled to have higher-resolution intermediate feature maps. From which, representative features are then grouped based on local pairwise feature similarity, to produce finer-grained, more precise attention maps highlighting task-relevant parts of the input. The obtained attention maps are ranked according to the activity level of the compound feature, which provides information regarding the important level of the highlighted regions. The proposed model can be easily adapted in a wide variety of modern deep models, where classification is involved. Extensive quantitative and qualitative experiments showcase more comprehensive and accurate visual explanations compared to state-of-the-art attention models and visualizations methods across multiple tasks including fine-grained image classification, few-shot classification, and person re-identification, without compromising the classification accuracy. The proposed visualization model sheds imperative light on how neural networks `pay their attention' differently in different tasks.

New fashion product sales forecasting is a challenging problem that involves many business dynamics and cannot be solved by classical forecasting approaches. In this paper, we investigate the effectiveness of systematically probing exogenous knowledge in the form of Google Trends time series and combining it with multi-modal information related to a brand-new fashion item, in order to effectively forecast its sales despite the lack of past data. In particular, we propose a neural network-based approach, where an encoder learns a representation of the exogenous time series, while the decoder forecasts the sales based on the Google Trends encoding and the available visual and metadata information. Our model works in a non-autoregressive manner, avoiding the compounding effect of large first-step errors. As a second contribution, we present VISUELLE, a publicly available dataset for the task of new fashion product sales forecasting, containing multimodal information for 5577 real, new products sold between 2016-2019 from Nunalie, an Italian fast-fashion company. The dataset is equipped with images of products, metadata, related sales, and associated Google Trends. We use VISUELLE to compare our approach against state-of-the-art alternatives and several baselines, showing that our neural network-based approach is the most accurate in terms of both percentage and absolute error. It is worth noting that the addition of exogenous knowledge boosts the forecasting accuracy by 1.5% WAPE wise, revealing the importance of exploiting informative external information. The code and dataset are both available at https://github.com/HumaticsLAB/GTM-Transformer.

This study investigates the impact of machine learning models on the generation of counterfactual explanations by conducting a benchmark evaluation over three different types of models: decision-tree (fully transparent, interpretable, white-box model), a random forest (a semi-interpretable, grey-box model), and a neural network (a fully opaque, black-box model). We tested the counterfactual generation process using four algorithms (DiCE, WatcherCF, prototype, and GrowingSpheresCF) in the literature in five different datasets (COMPAS, Adult, German, Diabetes, and Breast Cancer). Our findings indicate that: (1) Different machine learning models have no impact on the generation of counterfactual explanations; (2) Counterfactual algorithms based uniquely on proximity loss functions are not actionable and will not provide meaningful explanations; (3) One cannot have meaningful evaluation results without guaranteeing plausibility in the counterfactual generation process. Algorithms that do not consider plausibility in their internal mechanisms will lead to biased and unreliable conclusions if evaluated with the current state-of-the-art metrics; (4) A qualitative analysis is strongly recommended (together with a quantitative analysis) to ensure a robust analysis of counterfactual explanations and the potential identification of biases.

This paper focuses on efficient landmark management in radar based simultaneous localization and mapping (SLAM). Landmark management is necessary in order to maintain a consistent map of the estimated landmarks relative to the estimate of the platform's pose. This task is particularly important when faced with multiple detections from the same landmark and/or dynamic environments where the location of a landmark can change. A further challenge with radar data is the presence of false detections. Accordingly, we propose a simple yet efficient rule based solution for radar SLAM landmark management. Assuming a low-dynamic environment, there are several steps in our solution: new landmarks need to be detected and included, false landmarks need to be identified and removed, and the consistency of the landmarks registered in the map needs to be maintained. To illustrate our solution, we run an extended Kalman filter SLAM algorithm in an environment containing both stationary and temporally stationary landmarks. Our simulation results demonstrate that the proposed solution is capable of reliably managing landmarks even when faced with false detections and multiple detections from the same landmark.

Named entity recognition (NER) is the task to detect and classify the entity spans in the text. When entity spans overlap between each other, this problem is named as nested NER. Span-based methods have been widely used to tackle the nested NER. Most of these methods will get a score $n \times n$ matrix, where $n$ means the length of sentence, and each entry corresponds to a span. However, previous work ignores spatial relations in the score matrix. In this paper, we propose using Convolutional Neural Network (CNN) to model these spatial relations in the score matrix. Despite being simple, experiments in three commonly used nested NER datasets show that our model surpasses several recently proposed methods with the same pre-trained encoders. Further analysis shows that using CNN can help the model find more nested entities. Besides, we found that different papers used different sentence tokenizations for the three nested NER datasets, which will influence the comparison. Thus, we release a pre-processing script to facilitate future comparison.

NASA's tiny CAPSTONE spacecraft has run into trouble on its way to the moon and is currently tumbling out of control. The US space agency's $30 million probe, which is around the size of a microwave oven and weighs just 55 pounds, has also been experiencing temperature issues and had problems generating power from its solar panels. Toward the end of a major engine burn last Thursday (September 9), CAPSTONE experienced an anomaly that put the probe in a protective'safe mode', mission team members said. In an update issued this week, Advanced Space – the company that is managing the project for NASA – described it as a'dynamic operational situation'. It is not the first time CAPSTONE has hit a snag.

The results of our survey of more than 1,300 business leaders and 3,000 consumers globally suggest that establishing trust in products and experiences that leverage AI, digital technologies, and data not only meets consumer expectations but also could promote growth. The research indicates that organizations that are best positioned to build digital trust are also more likely than others to see annual growth rates of at least 10 percent on their top and bottom lines. However, only a small contingent of companies surveyed are set to deliver. The research suggests what these companies are doing differently. A majority of consumers believe that the companies they do business with provide the foundational elements of digital trust, which we define as confidence in an organization to protect consumer data, enact effective cybersecurity, offer trustworthy AI-powered products and services, and provide transparency around AI and data usage.

Most existing slot filling models tend to memorize inherent patterns of entities and corresponding contexts from training data. However, these models can lead to system failure or undesirable outputs when being exposed to spoken language perturbation or variation in practice. We propose a perturbed semantic structure awareness transferring method for training perturbation-robust slot filling models. Specifically, we introduce two MLM-based training strategies to respectively learn contextual semantic structure and word distribution from unsupervised language perturbation corpus. Then, we transfer semantic knowledge learned from upstream training procedure into the original samples and filter generated data by consistency processing. These procedures aim to enhance the robustness of slot filling models. Experimental results show that our method consistently outperforms the previous basic methods and gains strong generalization while preventing the model from memorizing inherent patterns of entities and contexts.

We consider the problem of estimating an upper bound on the capacity of a memoryless channel with unknown channel law and continuous output alphabet. A novel data-driven algorithm is proposed that exploits the dual representation of capacity where the maximization over the input distribution is replaced with a minimization over a reference distribution on the channel output. To efficiently compute the required divergence maximization between the conditional channel and the reference distribution, we use a modified mutual information neural estimator that takes the channel input as an additional parameter. We numerically evaluate our approach on different memoryless channels and show empirically that the estimated upper bounds closely converge either to the channel capacity or to best-known lower bounds.

Deep learning-based techniques have been introduced into the field of trajectory optimization in recent years. Deep Neural Networks (DNNs) are trained and used as the surrogates of conventional optimization process. They can provide low thrust (LT) transfer cost estimation and enable more complex preliminary mission designs. However, it is a challenge to efficiently obtain the required amount of trajectory data for training. A Generative Adversarial Network (GAN) is adapted to generate the feasible LT trajectory data efficiently. The GAN consists of a generator and a discriminator, both of which are deep networks. The generator generates fake LT transfer features using random noise as input, while the discriminator distinguishes the generator's fake LT transfer features from real LT transfer features. The GAN is trained until the generator generates fake LT transfers that the discriminator cannot identify. This indicates the generator generates low thrust transfer features that have the same distribution as the real transfer features. The generated low thrust transfer data have a high convergence rate, and they can be used to efficiently produce training data for deep learning models. The proposed approach is validated by generating feasible LT transfers in a Near-Earth Asteroid (NEA) mission scenario. The convergence rate of GAN-generated samples is 84.3%.