Deleting data from a trained machine learning (ML) model is a critical task in many applications. For example, we may want to remove the influence of training points that might be out of date or outliers. Regulations such as EU's General Data Protection Regulation also stipulate that individuals can request to have their data deleted. The naive approach to data deletion is to retrain the ML model on the remaining data, but this is too time consuming. Moreover there is no known efficient algorithm that exactly deletes data from most ML models. In this work, we evaluate several approaches for approximate data deletion from trained models. For the case of linear regression, we propose a new method with linear dependence on the feature dimension $d$, a significant gain over all existing methods which all have superlinear time dependence on the dimension. We also provide a new test for evaluating data deletion from linear models.

To effectively enforce fairness constraints one needs to define an appropriate notion of fairness and employ representation learning in order to impose this notion without compromising downstream utility for the data consumer. A desirable notion is individual fairness as it guarantees similar treatment for similar individuals. In this work, we introduce the first method which generalizes individual fairness to rich similarity notions via logical constraints while also enabling data consumers to obtain fairness certificates for their models. The key idea is to learn a representation that provably maps similar individuals to latent representations at most $\epsilon$ apart in $\ell_{\infty}$-distance, enabling data consumers to certify individual fairness by proving $\epsilon$-robustness of their classifier. Our experimental evaluation on six real-world datasets and a wide range of fairness constraints demonstrates that our approach is expressive enough to capture similarity notions beyond existing distance metrics while scaling to realistic use cases.

In a recent post, we described what it would take to build a sustainable machine learning practice. By "sustainable," we mean projects that aren't just proofs of concepts or experiments. A sustainable practice means projects that are integral to an organization's mission: projects by which an organization lives or dies. These projects are built and supported by a stable team of engineers, and supported by a management team that understands what machine learning is, why it's important, and what it's capable of accomplishing. Finally, sustainable machine learning means that as many aspects of product development as possible are automated: not just building models, but cleaning data, building and managing data pipelines, testing, and much more. Machine learning will penetrate our organizations so deeply that it won't be possible for humans to manage them unassisted. Organizations throughout the world are waking up to the fact that security is essential to their software projects. Nobody wants to be the next Sony, the next Anthem, or the next Equifax. But while we know how to make traditional software more secure (even though we frequently don't), machine learning presents a new set of problems. Any sustainable machine learning practice must address machine learning's unique security issues. We didn't do that for traditional software, and we're paying the price now.

There is increasing use of algorithms in the health care and criminal justice systems, and corresponding increased concern with their ethical use. But perhaps a more basic issue is whether we should believe what we hear about them and what the algorithm tells us. It is illuminating to distinguish between the trustworthiness of claims made about an algorithm, and those made by an algorithm, which reveals the potential contribution of statistical science to both evaluation and'intelligent transparency.' In particular, a four-phase evaluation structure is proposed, parallel to that adopted for pharmaceuticals. When on holiday in Portugal last year, we came to rely on'Mrs.

Artificial Intelligence technologies carrying a high-risk of abuse that could potentially lead to an erosion of fundamental rights will be subjected to a series of new requirements, the European Commission announced on Wednesday (19 February). As part of the executive's White paper on AI, a series of'high-risk' technologies have been earmarked for future oversight, including those in'critical sectors' and those deemed to be of'critical use.' Those under the critical sectors remit include healthcare, transport, police, recruitment, and the legal system, while technologies of critical use include such technologies with a risk of death, damage or injury, or with legal ramifications. Artificial Intelligence technologies coming under those two categories will be obliged to abide by strict rules, which could include compliance tests and controls, the Commission said on Wednesday. Sanctions could be imposed should certain technologies fail to meet such requirements.

The European Union has published its European data strategy [PDF], intended to provide the framework for what it describes as human-centric artificial intelligence. The white paper, said President of the European Commission, Ursula von der Leyen, is intended to "shape Europe's digital future". She continued: "It covers everything from cybersecurity to critical infrastructures, digital education to skills, democracy to media. I want that digital Europe reflects the best of Europe - open, fair, diverse, democratic, and confident." However, while the strategy has been pitched as boosting the EU's technology sector and preparing the bloc for a shift to an ever-more data-driven economy, increasingly governed by AI, the strategy is driven by a desire to regulate artificial intelligence and data platforms before they take off.

Tesla CEO Elon Musk wants to see all artificial intelligence better regulated, even at his own company, he tweeted Monday (via TechCrunch). He made the remark in response to a piece about OpenAI by MIT Technology Review, which claimed that the AI organization, co-founded by Musk, has shifted from its mission of developing and distributing AI safely and equitably into a secretive company obsessed with image and driven to constantly raise more money. Musk has a history of expressing serious concerns about the negative potential of AI. He tweeted in 2014 that it could be "more dangerous than nukes," and told an audience at an MIT Aeronautics and Astronautics symposium that year that AI was "our biggest existential threat," and humanity needs to be extremely careful: With artificial intelligence we are summoning the demon. In all those stories where there's the guy with the pentagram and the holy water, it's like yeah he's sure he can control the demon.

Oliver stated: "Whether businesses in different countries have different regulations or rules, or whether some individuals understand the use of their data more than others, I don't think individuals will ever be able to know and be able to give an informed version of consent for the use of their data and how it is then processed within AI. I don't think the current regulatory environment, even contemplates in a true sense, the effect of intelligent systems. Therefore, ethics and morality clauses should be included into businesses' terms of service and privacy policies."

The influence of human judgement is ubiquitous in datasets used across the analytics industry, yet humans are known to be sub-optimal decision makers prone to various biases. Analysing biased datasets then leads to biased outcomes of the analysis. Bias by protected characteristics (e.g. race) is of particular interest as it may not only make the output of analytical process sub-optimal, but also illegal. Countering the bias by constraining the analytical outcomes to be fair is problematic because A) fairness lacks a universally accepted definition, while at the same time some definitions are mutually exclusive, and B) the use of optimisation constraints ensuring fairness is incompatible with most analytical pipelines. Both problems are solved by methods which remove bias from the data and returning an altered dataset. This approach aims to not only remove the actual bias variable (e.g. race), but also alter all proxy variables (e.g. postcode) so the bias variable is not detectable from the rest of the data. The advantage of using this approach is that the definition of fairness as a lack of detectable bias in the data (as opposed to the output of analysis) is universal and therefore solves problem (A). Furthermore, as the data is altered to remove bias the problem (B) disappears because the analytical pipelines can remain unchanged. This approach has been adopted by several technical solutions. None of them, however, seems to be satisfactory in terms of ability to remove multivariate, non-linear and non-binary biases. Therefore, in this paper I propose the concept of Fair Adversarial Networks as an easy-to-implement general method for removing bias from data. This paper demonstrates that Fair Adversarial Networks achieve this aim.

The new goal is to seek high levels of human control AND high levels of automation, which is more likely to produce computer applications that are Reliable, Safe & Trustworthy (RST). Achieving this goal, especially for complex poorly understood problems, will dramatically increase human performance, while supporting human self-efficacy, mastery, creativity, and responsibility. The traditional belief in computer autonomy is compelling for many artificial intelligence (AI) researchers, developers, journalists, and promoters. The goal of computer autonomy was central in Sheridan and Verplank's (1978) ten levels from human control to computer automation/autonomy (Table 1). Their widely cited one-dimensional list continues to guide much of the research and development, suggesting that increases in automation must come at the cost of lowering human control. Shifting to HCAI could liberate design thinking so as to produce computer applications that increase automation, while amplifying, augmenting, enhancing, and empowering people to innovatively apply systems and creatively refine them.