Childcare is the most intimate of activities. Evolution has generated drives so powerful that we will risk our lives to protect not only our own children, but quite often any child, and even the young of other species. Robots, by contrast, are products created by commercial entities with commercial goals, which may--and should--include the well-being of their customers, but will never be limited to such. Robots, corporations, and other legal or non-legal entities do not possess the instinctual nature of humans to care for the young--even if our anthropomorphic tendencies may prompt some children and adults to overlook this fact. If you buy something using links in our stories, we may earn a commission.

In December 2020, we published a list of 10 predictions about the world of artificial intelligence in the year 2021. With 2021 now coming to a close, let's revisit these predictions to see how things actually played out. There is much to learn from these retrospectives about the state and trajectory of AI today. As of the beginning of this year, no autonomous vehicle company had ever gone public. TuSimple, Embark and Aurora have all debuted on public markets this year.

It may be smart, but it's not that clever. Artificial intelligence is nothing without human input. The algorithms that drive AI rely on the expertise of programmers and it's still no more than a tool – albeit a powerful one – that scientists and engineers can use to solve problems. Yet this is not to say that AI isn't the fastest-growing deep technology in the world, with the potential to transform people's lives and boost nations' economies. Facilitating AI innovation has even become a priority for the UK government, as laid out in the National AI Strategy it published in September.

That's because machine learning and artificial intelligence are assistive technologies that supplement (not replace) an attorney's role.

This paper addresses and solves some challenges in the adoption of machine learning in insurance with the democratization of model deployment. The first challenge is reducing the labelling effort (hence focusing on the data quality) with the help of active learning, a feedback loop between the model inference and an oracle: as in insurance the unlabeled data is usually abundant, active learning can become a significant asset in reducing the labelling cost. For that purpose, this paper sketches out various classical active learning methodologies before studying their empirical impact on both synthetic and real datasets. Another key challenge in insurance is the fairness issue in model inferences. We will introduce and integrate a post-processing fairness for multi-class tasks in this active learning framework to solve these two issues. Finally numerical experiments on unfair datasets highlight that the proposed setup presents a good compromise between model precision and fairness.

We provide a regularization framework for subject transfer learning in which we seek to train an encoder and classifier to minimize classification loss, subject to a penalty measuring independence between the latent representation and the subject label. We introduce three notions of independence and corresponding penalty terms using mutual information or divergence as a proxy for independence. For each penalty term, we provide several concrete estimation algorithms, using analytic methods as well as neural critic functions. We provide a hands-off strategy for applying this diverse family of regularization algorithms to a new dataset, which we call "AutoTransfer". We evaluate the performance of these individual regularization strategies and our AutoTransfer method on EEG, EMG, and ECoG datasets, showing that these approaches can improve subject transfer learning for challenging real-world datasets. In this work, we investigate methods for transfer learning in the classification of biosignals data. Previous work has established the difficulty of transfer learning for biosignals and even the issue of so-called "negative transfer", in which naive attempts to combine datasets from multiple subjects or sessions can paradoxically decrease model performance, due to differences in response statistics [1, 2]. We address the problem of subject transfer by training models to be invariant to changes in a nuisance variable representing subject identifier.

When legal regulations get ahead of technological developments, the progress of human society may be constrained. When technological developments run ahead of legal regulations, the unregulated new technologies may harm instead of benefit human society, defying technological development's fundamental purpose. This is exactly what has happened in our world in the past decade, as technological developments far outpaced legal regulations. Worse, traditional legal frameworks focus on the relation between people, whereas we must develop a legal framework to regulate relations between people and intelligent machines in the current era. Integrating AI technologies into human society imposes unique legal challenges without any precedence.

A 2013 article from NPR's "Code Switch," which explores issues of race and identity, delved into the etymology of the term after it surfaced in George Zimmerman's trial for the killing of Trayvon Martin. Academics and historians interviewed by the article's author, Gene Demby, dated the term's use back to Shakespearean times when it was applied as an "insult for an obnoxious bloviator" and was usually directed at people from Scotland or Ireland. When immigrants from those countries crossed the Atlantic to America, the term followed them. Jelani Cobb, a historian now at Columbia University and a staff writer for the New Yorker interviewed by Demby, noted it was later tied to poor White farm hands "since the manual labor they did involved driving livestock with a whip."

In his 2021 Reith lectures, the third episode of which airs tonight, the artificial intelligence researcher Stuart Russell takes up the idea of a near-future AI that is so ruthlessly intelligent that it might pose an existential threat to humanity. A machine we create that might destroy us all. This has long been a popular topic with researchers and the press. But we believe an existential threat from AI is both unlikely and in any case far off, given the current state of the technology. However, the recent development of powerful, but far smaller-scale, AI systems has had a significant effect on the world already, and the use of existing AI poses serious economic and social challenges.

From looking into waters for their reflection to mirrors and then cameras, we, humans, have come a long way. After digital cameras came into existence, it was possible to have databases of large numbers of faces and their facial features. The advancement of software and technologies made it possible to use this database, run for facial recognition technology. It is now capable of analyzing and recognizing human faces as accurately as possible.