One of the difficulties of artificial intelligence is to ensure that model decisions are fair and free of bias. In research, datasets, metrics, techniques, and tools are applied to detect and mitigate algorithmic unfairness and bias. This study examines the current knowledge on bias and unfairness in machine learning models. The systematic review followed the PRISMA guidelines and is registered on OSF plataform. The search was carried out between 2021 and early 2022 in the Scopus, IEEE Xplore, Web of Science, and Google Scholar knowledge bases and found 128 articles published between 2017 and 2022, of which 45 were chosen based on search string optimization and inclusion and exclusion criteria. We discovered that the majority of retrieved works focus on bias and unfairness identification and mitigation techniques, offering tools, statistical approaches, important metrics, and datasets typically used for bias experiments. In terms of the primary forms of bias, data, algorithm, and user interaction were addressed in connection to the preprocessing, in-processing, and postprocessing mitigation methods. The use of Equalized Odds, Opportunity Equality, and Demographic Parity as primary fairness metrics emphasizes the crucial role of sensitive attributes in mitigating bias. The 25 datasets chosen span a wide range of areas, including criminal justice image enhancement, finance, education, product pricing, and health, with the majority including sensitive attributes. In terms of tools, Aequitas is the most often referenced, yet many of the tools were not employed in empirical experiments. A limitation of current research is the lack of multiclass and multimetric studies, which are found in just a few works and constrain the investigation to binary-focused method. Furthermore, the results indicate that different fairness metrics do not present uniform results for a given use case, and that more research with varied model architectures is necessary to standardize which ones are more appropriate for a given context. We also observed that all research addressed the transparency of the algorithm, or its capacity to explain how decisions are taken.

Head-mounted displays are virtual reality devices that may be equipped with sensors and cameras to measure a patient’s heart rate through facial regions. Heart rate is an essential body signal that can be used to remotely monitor users in a variety of situations. There is currently no study that predicts heart rate using only highlighted facial regions; thus, an adaptation is required for beats per minute predictions. Likewise, there are no datasets containing only the eye and lower face regions, necessitating the development of a simulation mechanism. This work aims to remotely estimate heart rate from facial regions that can be captured by the cameras of a head-mounted display using state-of-the-art EVM-CNN and Meta-rPPG techniques. We developed a region of interest extractor to simulate a dataset from a head-mounted display device using stabilizer and video magnification techniques. Then, we combined support vector machine and FaceMash to determine the regions of interest and adapted photoplethysmography and beats per minute signal predictions to work with the other techniques. We observed an improvement of 188.88% for the EVM and 55.93% for the Meta-rPPG. In addition, both models were able to predict heart rate using only facial regions as input. Moreover, the adapted technique Meta-rPPG outperformed the original work, whereas the EVM adaptation produced comparable results for the photoplethysmography signal.

Generally, the action recognition task requires a vast amount of labeled data, which represents a time-consuming human annotation effort. To mitigate the dependency on labeled data, this study proposes Semi-Supervised and Iterative Reinforcement Learning (RL-SSI), which adapts a supervised approach that uses 100% labeled data to a semi-supervised and iterative approach using reinforcement learning for human action recognition in videos. The JIGSAWS and Breakfast datasets were used to evaluate the RL-SSI model, because they are commonly used in the action segmentation task. The same applies to the performance metrics used in this work-F-Score (F1) and Edit Score-which are commonly applied for such tasks. In JIGSAWS tests, we observed that the RL-SSI outperformed previously developed state-of-the-art techniques in all quantitative measures, while using only 65% of the labeled data. When analysing the Breakfast tests, we compared the effectiveness of RL-SSI with the results of the self-supervised technique called SSTDA. We have found that RL-SSI outperformed SSTDA with an accuracy of 66.44% versus 65.8%, but RL-SSI was surpassed by the F1@10 segmentation measure, which presented an accuracy of 67.33% versus 69.3% for SSTDA. Despite this, our experiment only used 55.8% of the labeled data, while SSTDA used 65%. We conclude that our approach outperformed equivalent supervised learning methods and is comparable to SSTDA, when evaluated on multiple datasets of human action recognition, proving to be an important innovative method to successfully building solutions to reduce the amount of fully labeled data, leveraging the work of human specialists in the task of data labeling of videos, and their respectives frames, for human action recognition, thus reducing the required resources to accomplish it.