In object detection, false negatives arise when a detector fails to detect a target object. To understand why object detectors produce false negatives, we identify five 'false negative mechanisms', where each mechanism describes how a specific component inside the detector architecture failed. Focusing on two-stage and one-stage anchor-box object detector architectures, we introduce a framework for quantifying these false negative mechanisms. Using this framework, we investigate why Faster R-CNN and RetinaNet fail to detect objects in benchmark vision datasets and robotics datasets. We show that a detector's false negative mechanisms differ significantly between computer vision benchmark datasets and robotics deployment scenarios. This has implications for the translation of object detectors developed for benchmark datasets to robotics applications. Code is publicly available at https://github.com/csiro-robotics/fn_mechanisms

The boll weevil (Anthonomus grandis L.) is a serious pest that primarily feeds on cotton plants. In places like Lower Rio Grande Valley of Texas, due to sub-tropical climatic conditions, cotton plants can grow year-round and therefore the left-over seeds from the previous season during harvest can continue to grow in the middle of rotation crops like corn (Zea mays L.) and sorghum (Sorghum bicolor L.). These feral or volunteer cotton (VC) plants when reach the pinhead squaring phase (5-6 leaf stage) can act as hosts for the boll weevil pest. The Texas Boll Weevil Eradication Program (TBWEP) employs people to locate and eliminate VC plants growing by the side of roads or fields with rotation crops but the ones growing in the middle of fields remain undetected. In this paper, we demonstrate the application of computer vision (CV) algorithm based on You Only Look Once version 5 (YOLOv5) for detecting VC plants growing in the middle of corn fields at three different growth stages (V3, V6, and VT) using unmanned aircraft systems (UAS) remote sensing imagery. All the four variants of YOLOv5 (s, m, l, and x) were used and their performances were compared based on classification accuracy, mean average precision (mAP), and F1-score. It was found that YOLOv5s could detect VC plants with a maximum classification accuracy of 98% and mAP of 96.3 % at the V6 stage of corn while YOLOv5s and YOLOv5m resulted in the lowest classification accuracy of 85% and YOLOv5m and YOLOv5l had the least mAP of 86.5% at the VT stage on images of size 416 x 416 pixels. The developed CV algorithm has the potential to effectively detect and locate VC plants growing in the middle of corn fields as well as expedite the management aspects of TBWEP.

DNNs are known to be vulnerable to so-called adversarial attacks that manipulate inputs to cause incorrect results that can be beneficial to an attacker or damaging to the victim. Recent works have proposed approximate computation as a defense mechanism against machine learning attacks. We show that these approaches, while successful for a range of inputs, are insufficient to address stronger, high-confidence adversarial attacks. To address this, we propose DNNSHIELD, a hardware-accelerated defense that adapts the strength of the response to the confidence of the adversarial input. Our approach relies on dynamic and random sparsification of the DNN model to achieve inference approximation efficiently and with fine-grain control over the approximation error. DNNSHIELD uses the output distribution characteristics of sparsified inference compared to a dense reference to detect adversarial inputs. We show an adversarial detection rate of 86% when applied to VGG16 and 88% when applied to ResNet50, which exceeds the detection rate of the state of the art approaches, with a much lower overhead. We demonstrate a software/hardware-accelerated FPGA prototype, which reduces the performance impact of DNNSHIELD relative to software-only CPU and GPU implementations.

In this new era of social media, social networks are becoming increasingly important sources of user-generated content on the internet. These kinds of information resources, which include a lot of people's feelings, opinions, feedback, and reviews, are very useful for big businesses, markets, politics, journalism, and many other fields. Politics is one of the most talked-about and popular topics on social media networks right now. Many politicians use micro-blogging services like Twitter because they have a large number of followers and supporters on those networks. Politicians, political parties, political organizations, and foundations use social media networks to communicate with citizens ahead of time. Today, social media is used by hundreds of thousands of political groups and politicians. On these social media networks, every politician and political party has millions of followers, and politicians find new and innovative ways to urge individuals to participate in politics. Furthermore, social media assists politicians in various decision-making processes by providing recommendations, such as developing policies and strategies based on previous experiences, recommending and selecting suitable candidates for a particular constituency, recommending a suitable person for a particular position in the party, and launching a political campaign based on citizen sentiments on various issues and controversies, among other things. This research is a review on the use of social network analysis (SNA) and semantic analysis (SA) on the Twitter platform to study the supporters networks of political leaders because it can help in decision-making when predicting their political futures.

This paper proposes a novel out-of-distribution (OOD) detection framework named MoodCat for image classifiers. MoodCat masks a random portion of the input image and uses a generative model to synthesize the masked image to a new image conditioned on the classification result. It then calculates the semantic difference between the original image and the synthesized one for OOD detection. Compared to existing solutions, MoodCat naturally learns the semantic information of the in-distribution data with the proposed mask and conditional synthesis strategy, which is critical to identifying OODs. Experimental results demonstrate that MoodCat outperforms state-of-the-art OOD detection solutions by a large margin.

The management of hyperglycemia in hospitalized patients has a significant impact on both morbidity and mortality. Therefore, it is important to predict the need for diabetic patients to be hospitalized. However, using standard machine learning approaches to make these predictions may result in health disparities caused by biases in the data related to social determinants (such as race, age, and gender). These biases must be removed early in the data collection process, before they enter the system and are reinforced by model predictions, resulting in biases in the model's decisions. In this paper, we propose a machine learning pipeline capable of making predictions as well as detecting and mitigating biases in the data and model predictions. This pipeline analyses the clinical data and determines whether biases exist in the data, if so, it removes those biases before making predictions. We evaluate the performance of the proposed method on a clinical dataset using accuracy and fairness measures. The findings of the results show that when we mitigate biases early during the data ingestion, we get fairer predictions.

Educational technologies nowadays increasingly use data and Machine Learning (ML) models. This gives the students, instructors, and administrators support and insights for the optimum policy. However, it is well acknowledged that ML models are subject to bias, which raises concerns about the fairness, bias, and discrimination of using these automated ML algorithms in education and its unintended and unforeseen negative consequences. The contribution of bias during the decision-making comes from datasets used for training ML models and the model architecture. This paper presents a preliminary investigation of the fairness of transformer neural networks on the two tabular datasets: Law School and Student-Mathematics. In contrast to classical ML models, the transformer-based models transform these tabular datasets into a richer representation while solving the classification task. We use different fairness metrics for evaluation and check the trade-off between fairness and accuracy of the transformer-based models over the tabular datasets. Empirically, our approach shows impressive results regarding the trade-off between fairness and performance on the Law School dataset.

In human-robot collaboration, robot errors are inevitable -- damaging user trust, willingness to work together, and task performance. Prior work has shown that people naturally respond to robot errors socially and that in social interactions it is possible to use human responses to detect errors. However, there is little exploration in the domain of non-social, physical human-robot collaboration such as assembly and tool retrieval. In this work, we investigate how people's organic, social responses to robot errors may be used to enable timely automatic detection of errors in physical human-robot interactions. We conducted a data collection study to obtain facial responses to train a real-time detection algorithm and a case study to explore the generalizability of our method with different task settings and errors. Our results show that natural social responses are effective signals for timely detection and localization of robot errors even in non-social contexts and that our method is robust across a variety of task contexts, robot errors, and user responses. This work contributes to robust error detection without detailed task specifications.

Legged robots can traverse a wide variety of terrains, some of which may be challenging for wheeled robots, such as stairs or highly uneven surfaces. However, quadruped robots face stability challenges on slippery surfaces. This can be resolved by adjusting the robot's locomotion by switching to more conservative and stable locomotion modes, such as crawl mode (where three feet are in contact with the ground always) or amble mode (where one foot touches down at a time) to prevent potential falls. To tackle these challenges, we propose an approach to learn a model from past robot experience for predictive detection of potential failures. Accordingly, we trigger gait switching merely based on proprioceptive sensory information. To learn this predictive model, we propose a semi-supervised process for detecting and annotating ground truth slip events in two stages: We first detect abnormal occurrences in the time series sequences of the gait data using an unsupervised anomaly detector, and then, the anomalies are verified with expert human knowledge in a replay simulation to assert the event of a slip. These annotated slip events are then used as ground truth examples to train an ensemble decision learner for predicting slip probabilities across terrains for traversability. We analyze our model on data recorded by a legged robot on multiple sites with slippery terrain. We demonstrate that a potential slip event can be predicted up to 720 ms ahead of a potential fall with an average precision greater than 0.95 and an average F-score of 0.82. Finally, we validate our approach in real-time by deploying it on a legged robot and switching its gait mode based on slip event detection.

Studies show that Studies that cardiovascular diseases (CVDs) are malignant for human health. Thus, it is important to have an efficient way of CVD prognosis. In response to this, the healthcare industry has adopted machine learning-based smart solutions to alleviate the manual process of CVD prognosis. Thus, this work proposes an information fusion technique that combines key attributes of a person through analysis of variance (ANOVA) and domain experts' knowledge. It also introduces a new collection of CVD data samples for emerging research. There are thirty-eight experiments conducted exhaustively to verify the performance of the proposed framework on four publicly available benchmark datasets and the newly created dataset in this work. The ablation study shows that the proposed approach can achieve a competitive mean average accuracy (mAA) of 99.2% and a mean average AUC of 97.9%.