Unlikepriorworkon individual fairness, we do not assume the similarity measure among individuals is known, nor do we assume that such measure takes a certain parametric form.

Inc (AMZN.O) is planning to roll out warning cards for software sold by its cloud-computing division, in light of ongoing concern that artificially intelligent systems can discriminate against different groups, the company told Reuters. Akin to lengthy nutrition labels, Amazon's so-called AI Service Cards will be public so its business customers can see the limitations of certain cloud services, such as facial recognition and audio transcription. The goal would be to prevent mistaken use of its technology, explain how its systems work and manage privacy, Amazon said. The company is not the first to publish such warnings. International Business Machines Corp (IBM.N), a smaller player in the cloud, did so years ago.

Emphasizing issues of computational efficiency, Michael Kearns and Umesh Vazirani introduce a number of central topics in computational learning theory for researchers and students in artificial intelligence, neural networks, theoretical computer science, and statistics. Computational learning theory is a new and rapidly expanding area of research that examines formal models of induction with the goals of discovering the common methods underlying efficient learning algorithms and identifying the computational impediments to learning. Each topic in the book has been chosen to elucidate a general principle, which is explored in a precise formal setting. Intuition has been emphasized in the presentation to make the material accessible to the nontheoretician while still providing precise arguments for the specialist. This balance is the result of new proofs of established theorems, and new presentations of the standard proofs.

The frequency and magnitude of natural disasters such as hurricanes, wildfires and floods has been growing for a number of years. Panelists for the Advanced Technology Academic Research Center's Jan. 26 webinar "Leveraging Predictive Analytics to Address Climate Change Issues" discussed how the use of artificial intelligence and machine learning can provide both short- and long-term guidance for decisionmakers considering how to ameliorate impacts. According to the National Centers for Environmental Information, which is part of the National Oceanic and Atmospheric Administration, in 2021 there were 20 weather and climate disaster events that incurred losses of more than $1 billion each. From 1980 to 2021, the annual average was 7.4 events (adjusted for inflation), but from 2017 to 2021, the most recent five years, the average number of events was 17.2 (adjusted for inflation). In short, the problem is getting worse, and faster.

This work provides several new insights on the robustness of Kearns' statistical query framework against challenging label-noise models. First, we build on a recent result by \cite{DBLP:journals/corr/abs-2006-04787} that showed noise tolerance of distribution-independently evolvable concept classes under Massart noise. Specifically, we extend their characterization to more general noise models, including the Tsybakov model which considerably generalizes the Massart condition by allowing the flipping probability to be arbitrarily close to $\frac{1}{2}$ for a subset of the domain. As a corollary, we employ an evolutionary algorithm by \cite{DBLP:conf/colt/KanadeVV10} to obtain the first polynomial time algorithm with arbitrarily small excess error for learning linear threshold functions over any spherically symmetric distribution in the presence of spherically symmetric Tsybakov noise. Moreover, we posit access to a stronger oracle, in which for every labeled example we additionally obtain its flipping probability. In this model, we show that every SQ learnable class admits an efficient learning algorithm with OPT + $\epsilon$ misclassification error for a broad class of noise models. This setting substantially generalizes the widely-studied problem of classification under RCN with known noise rate, and corresponds to a non-convex optimization problem even when the noise function -- i.e. the flipping probabilities of all points -- is known in advance.

The last decade has seen a vast increase both in the diversity of applications to which machine learning is applied, and to the import of those applications. Machine learning is no longer just the engine behind ad placements and spam filters; it is now used to filter loan applicants, deploy police officers, and inform bail and parole decisions, among other things. The result has been a major concern for the potential for data-driven methods to introduce and perpetuate discriminatory practices, and to otherwise be unfair. And this concern has not been without reason: a steady stream of empirical findings has shown that data-driven methods can unintentionally both encode existing human biases and introduce new ones.7,9,11,60 At the same time, the last two years have seen an unprecedented explosion in interest from the academic community in studying fairness and machine learning. "Fairness and transparency" transformed from a niche topic with a trickle of papers produced every year (at least since the work of Pedresh56 to a major subfield of machine learning, complete with a dedicated archival conference--ACM FAT*). But despite the volume and velocity of published work, our understanding of the fundamental questions related to fairness and machine learning remain in its infancy.

In 2009, Netflix was sued for releasing movie ratings data from half a million subscribers who were identified only by unique ID numbers. The video streaming service divulged this "anonymized" information to the public as part of its Netflix Prize contest, in which participants were asked to use the data to develop a better content recommendation algorithm. But researchers from the University of Texas showed that as few as six movie ratings could be used to identify users. A closet lesbian sued Netflix, saying her anonymity was compromised. The lawsuit was settled in 2010. The Netflix case reveals a problem about which the public is just starting to learn, but that data analysts and computer scientists have known for years. In anonymized datasets where distinguishing characteristics of a person such as name and address have been deleted, even a handful of seemingly innocuous information can lead to identification.

Kearns is a computer science professor at the University of Pennsylvania and has years of experience at Steve Cohen's former hedge fund and other Wall Street firms. He will lead Morgan Stanley's AI research and offer advice on deploying the technology for projects across the company, the New York-based firm said in a memo to employees Tuesday. Chief Executive Officer James Gorman has made new technologies a top priority at Morgan Stanley, an early mover in electronic stock trading. The bank is spending about $4 billion in an initiative that spans trading -- particularly in fixed income -- wealth management and other units. "Michael uniquely combines a distinguished career in academia and research with professional experience in the application of AI to complex business problems in financial services," the company said in the memo.