In an era of cancel-culture and one where racial discrimination is currently extremely socially embedded in the minds of the majority and more importantly of potential customers, brands cannot risk being seen to be discriminatory in any way possible. Being racist, sexist, homophobic, etc. now as a brand will lead to irreversible damage. The fear of this is profound for many because the experience of their AI-powered algorithms expressing bias against people of colour or against women is fresh in their minds. For instance, in 2014, Amazon discovered that an internally developed AI algorithm to automate headhunting, sold to recruitment firms and such to find potentially strong candidates, taught itself sexism. The AI taught itself that male applicants were more desirable than female applicants.

As we rely on machine learning (ML) models to make more consequential decisions, the issue of ML models perpetuating or even exacerbating undesirable historical biases (e.g. In this paper, we focus on the problem of detecting violations of individual fairness in ML models. We formalize the problem as measuring the susceptibility of ML models against a form of adversarial attack and develop a suite of inference tools for the adversarial cost function. The tools allow auditors to assess the individual fairness of ML models in a statistically-principled way: form confidence intervals for the worst-case performance differential between similar individuals and test hypotheses of model fairness with (asymptotic) non-coverage/Type I error rate control. The problem of bias in machine learning systems is at the forefront of contemporary ML research. Numerous media outlets have scrutinized machine learning systems deployed in practice for violations of basic societal equality principles (Angwin et al., 2016; Dastin, 2018; Vigdor, 2019). In response researchers developed many formal definitions of algorithmic fairness along with algorithms for enforcing these definitions in ML models (Dwork et al., 2011; Hardt et al., 2016; Berk et al., 2017; Kusner et al., 2018; Ritov et al., 2017; Yurochkin et al., 2020). Despite the flurry of ML fairness research, the basic question of assessing fairness of a given ML model in a statistically principled way remains largely unexplored. In this paper we propose a statistically principled approach to assessing individual fairness (Dwork et al., 2011) of ML models.

We consider the task of enforcing individual fairness in gradient boosting. Gradient boosting is a popular method for machine learning from tabular data, which arise often in applications where algorithmic fairness is a concern. At a high level, our approach is a functional gradient descent on a (distributionally) robust loss function that encodes our intuition of algorithmic fairness for the ML task at hand. Unlike prior approaches to individual fairness that only work with smooth ML models, our approach also works with non-smooth models such as decision trees. We show that our algorithm converges globally and generalizes. We also demonstrate the efficacy of our algorithm on three ML problems susceptible to algorithmic bias.

We provide a construction for categorical representation learning and introduce the foundations of "$\textit{categorifier}$". The central theme in representation learning is the idea of $\textbf{everything to vector}$. Every object in a dataset $\mathcal{S}$ can be represented as a vector in $\mathbb{R}^n$ by an $\textit{encoding map}$ $E: \mathcal{O}bj(\mathcal{S})\to\mathbb{R}^n$. More importantly, every morphism can be represented as a matrix $E: \mathcal{H}om(\mathcal{S})\to\mathbb{R}^{n}_{n}$. The encoding map $E$ is generally modeled by a $\textit{deep neural network}$. The goal of representation learning is to design appropriate tasks on the dataset to train the encoding map (assuming that an encoding is optimal if it universally optimizes the performance on various tasks). However, the latter is still a $\textit{set-theoretic}$ approach. The goal of the current article is to promote the representation learning to a new level via a $\textit{category-theoretic}$ approach. As a proof of concept, we provide an example of a text translator equipped with our technology, showing that our categorical learning model outperforms the current deep learning models by 17 times. The content of the current article is part of the recent US patent proposal (patent application number: 63110906).

Terence Mauri has won plaudits for his commentary on disruptive technology. He holds visiting positions at MIT and the London Business school, and his views are widely published. So when I saw an article about his predictions for the legal system I could be sure it would be thought-provoking. I wasn't disappointed: "Robotic judges that can determine guilt will be'commonplace' within 50 years"1. That's quite a claim, and as Niels Bohr quipped, "Prediction is very difficult, especially about the future…".

Sticks and stones may break my bones, but words will never hurt me. This was a mantra I picked up on the playground at elementary school--something I repeated over and over again anytime I came face to face with racism. It was a coping mechanism meant to guard my heart from the cacophony of discriminatory comments that shaped me as a young Korean American girl growing up in predominantly white spaces. But now that I'm well into adulthood, I think about the girls of color who are also being taught to pretend that words don't hurt--and the people this way of thinking actually protects. It's hard to escape the unrelenting consequences of racism: In the past year alone, we lost Breonna Taylor, George Floyd, Ahmaud Arbery, and the six women of Asian descent murdered in Atlanta (Xiaojie "Emily" Tan, Daoyou Feng, Suncha Kim, Yong Ae Yue, Soon Chung Park, Hyun Jung Grant) at the hands of this insidious disease--and those are just the names that were in the headlines. If we don't acknowledge ...

The future will involve artificial intelligence, but (in case you weren't already cautious) we need to be very careful what data we feed into these systems. A team led by computer scientists from MIT examined ten of the most-cited datasets used to test machine learning systems and found that around 3.4 percent of the data was inaccurate or mislabeled. Some have been cited over 100,000 times in machine learning research. Some of the errors are bigger than others, but include mistaking Bruce Springsteen for an orchestra and mislabelling a baby as a nipple. These errors could have huge ramifications for machine learning systems unless addressed. It may still be the early days of artificial intelligence, so we might want to check its homework.

The year is 2029, and you wake up one morning living in a community called Hope, a dystopian dictatorship. "Everyone here wears the same outfit, lives the same repetitive routine, and is happy … For many, Hope is their entire universe. They are uninterested in the outside world. However, you are different--you have the ability to choose." This is how you are introduced to the game Name of the Will on Kickstarter.

Machine learning and artificial intelligence have taken organizations to new heights of innovation, growth, and profits thanks to their ability to analyze data efficiently and with extreme accuracy. However, the inherent nature of some algorithms such as black-box models have been proven, at times, to be unfair and lack transparency, leading to multiplicated bias and detrimental impact on minorities. There are several key issues presented by black-box models, and they all work together to further bias data. The most prominent are models fed with data that is historically biased to begin with, and fed by humans who are biased by nature. In addition, because data analysts can only see the inputs and outputs but not the internal workings of how results are determined, machine learning is constantly aggregating this data, including personal data.

In this AI episode of Breaking Banks, JP Nicols hosts Alfred Cowger author of The Threats of Algorithms and A.I. to Civil Rights, Legal Remedies, and American Jurisprudence: One Nation Under Algorithms. JP and Al explore AI and algorithmic models and the intense policy debates surrounding the legality of, and liability for, these advanced machine learning applications. Then enjoy a special compilation of the best AI moments in Breaking Banks history. We begin with AI and Robots- Brett King hosts Ben Goertzel to learn about Sophia the social humanoid robot. We hear from Greg Cross of Soulmachines to learn how AI can be used for good and are digital humans the answer?