More Like this

1. Mitigating Dataset Harms Requires Stewardship: Lessons from 1000 Papers

   Kenny Peng, Arunesh Mathur (November 2021, Advances in neural information processing systems)

   Machine learning datasets have elicited concerns about privacy, bias, and unethical applications, leading to the retraction of prominent datasets such as DukeMTMC, MS-Celeb-1M, and Tiny Images. In response, the machine learning community has called for higher ethical standards in dataset creation. To help inform these efforts, we studied three influential but ethically problematic face and person recognition datasets—Labeled Faces in the Wild (LFW), MS-Celeb-1M, and DukeMTMC— by analyzing nearly 1000 papers that cite them. We found that the creation of derivative datasets and models, broader technological and social change, the lack of clarity of licenses, and dataset management practices can introduce a wide range of ethical concerns. We conclude by suggesting a distributed approach to harm mitigation that considers the entire life cycle of a dataset. 

   more » « less
2. Challenges and Benchmark Datasets for Machine Learning in the Atmospheric Sciences: Definition, Status, and Outlook

   https://doi.org/10.1175/AIES-D-21-0002.1  

   Dueben, Peter D.; Schultz, Martin G.; Chantry, Matthew; Gagne, David John; Hall, David Matthew; McGovern, Amy (July 2022, Artificial Intelligence for the Earth Systems)

   Abstract Benchmark datasets and benchmark problems have been a key aspect for the success of modern machine learning applications in many scientific domains. Consequently, an active discussion about benchmarks for applications of machine learning has also started in the atmospheric sciences. Such benchmarks allow for the comparison of machine learning tools and approaches in a quantitative way and enable a separation of concerns for domain and machine learning scientists. However, a clear definition of benchmark datasets for weather and climate applications is missing with the result that many domain scientists are confused. In this paper, we equip the domain of atmospheric sciences with a recipe for how to build proper benchmark datasets, a (nonexclusive) list of domain-specific challenges for machine learning is presented, and it is elaborated where and what benchmark datasets will be needed to tackle these challenges. We hope that the creation of benchmark datasets will help the machine learning efforts in atmospheric sciences to be more coherent, and, at the same time, target the efforts of machine learning scientists and experts of high-performance computing to the most imminent challenges in atmospheric sciences. We focus on benchmarks for atmospheric sciences (weather, climate, and air-quality applications). However, many aspects of this paper will also hold for other aspects of the Earth system sciences or are at least transferable. Significance Statement Machine learning is the study of computer algorithms that learn automatically from data. Atmospheric sciences have started to explore sophisticated machine learning techniques and the community is making rapid progress on the uptake of new methods for a large number of application areas. This paper provides a clear definition of so-called benchmark datasets for weather and climate applications that help to share data and machine learning solutions between research groups to reduce time spent in data processing, to generate synergies between groups, and to make tool developments more targeted and comparable. Furthermore, a list of benchmark datasets that will be needed to tackle important challenges for the use of machine learning in atmospheric sciences is provided. 

   more » « less
3. Exploring Generative Models with Middle School Students

   https://doi.org/10.1145/3411764.3445226  

   Ali, Safinah; DiPaola, Daniella; Lee, Irene; Hong, Jenna; Breazeal, Cynthia (May 2021, CHI '21: Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems)

   null (Ed.)

   Applications of generative models such as Generative Adversarial Networks (GANs) have made their way to social media platforms that children frequently interact with. While GANs are associated with ethical implications pertaining to children, such as the generation of Deepfakes, there are negligible efforts to educate middle school children about generative AI. In this work, we present a generative models learning trajectory (LT), educational materials, and interactive activities for young learners with a focus on GANs, creation and application of machine-generated media, and its ethical implications. The activities were deployed in four online workshops with 72 students (grades 5-9). We found that these materials enabled children to gain an understanding of what generative models are, their technical components and potential applications, and benefits and harms, while reflecting on their ethical implications. Learning from our findings, we propose an improved learning trajectory for complex socio-technical systems. 

   more » « less
4. Dimensions of UX Practice that Shape Ethical Awareness

   https://doi.org/10.1145/3313831.3376459  

   Chivukula, Shruthi Sai; Watkins, Chris Rhys; Manocha, Rhea; Chen, Jingle; Gray, Colin M. (April 2020, CHI’20: Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems)

   HCI researchers are increasingly interested in describing the complexity of design practice, including ethical, organizational, and societal concerns. Recent studies have identified individual practitioners as key actors in driving the design process and culture within their respective organizations, and we build upon these efforts to reveal practitioner concerns regarding ethics on their own terms. In this paper, we report on the results of an interview study with eleven UX practitioners, capturing their experiences that highlight dimensions of design practice that impact ethical awareness and action. Using a bottom-up thematic analysis, we identified five dimensions of design complexity that influence ethical outcomes and span individual, collaborative, and methodological framing of UX activity. Based on these findings, we propose a set of implications for the creation of ethically-centered design methods that resonate with this complexity and inform the education of future UX practitioners. 

   more » « less
5. Kn0w Thy Doma1n Name: Unbiased Phishing Detection Using Domain Name Based Features

   https://doi.org/10.1145/3205977.3205992  

   Shirazi, Hossein; Bezawada, Bruhadeshwar; Ray, Indrakshi (June 2018, Proceedings of the 23nd ACM on Symposium on Access Control Models and Technologies)

   Phishing websites remain a persistent security threat. Thus far, machine learning approaches appear to have the best potential as defenses. But, there are two main concerns with existing machine learning approaches for phishing detection. The first is the large number of training features used and the lack of validating arguments for these feature choices. The second concern is the type of datasets used in the literature that are inadvertently biased with respect to the features based on the website URL or content. To address these concerns, we put forward the intuition that the domain name of phishing websites is the tell-tale sign of phishing and holds the key to successful phishing detection. Accordingly, we design features that model the relationships, visual as well as statistical, of the domain name to the key elements of a phishing website, which are used to snare the end-users. The main value of our feature design is that, to bypass detection, an attacker will find it very difficult to tamper with the visual content of the phishing website without arousing the suspicion of the end user. Our feature set ensures that there is minimal or no bias with respect to a dataset. Our learning model trains with only seven features and achieves a true positive rate of 98% and a classification accuracy of 97%, on sample dataset. Compared to the state-of-the-art work, our per data instance classification is 4 times faster for legitimate websites and 10 times faster for phishing websites. Importantly, we demonstrate the shortcomings of using features based on URLs as they are likely to be biased towards specific datasets. We show the robustness of our learning algorithm by testing on unknown live phishing URLs and achieve a high detection accuracy of 99.7%. 

   more » « less