The U.S. Congress should create a federal agency for artificial intelligence. In 1906, in response to shocking reports about the disgusting conditions in U.S. meat-packing facilities, Congress created the Food and Drug Administration (FDA) to ensure safe and sanitary food production. In 1934, in the wake of the worst stock market crash in U.S. history, Congress created the Securities and Exchange Commission (SEC) to regulate capital markets. In 1970, as the nation became increasingly alarmed about the deterioration of the natural environment, Congress created the Environmental Protection Agency (EPA) to ensure cleaner skies and waters. When an entire field begins to create a broad set of challenges for the public, demanding thoughtful regulation, a proven governmental approach is to create a federal agency focused specifically on engaging with and managing that field.

We consider a resource-aware variant of the classical multi-armed bandit problem: In each round, the learner selects an arm and determines a resource limit. It then observes a corresponding (random) reward, provided the (random) amount of consumed resources remains below the limit. Otherwise, the observation is censored, i.e., no reward is obtained. For this problem setting, we introduce a measure of regret, which incorporates the actual amount of allocated resources of each learning round as well as the optimality of realizable rewards. Thus, to minimize regret, the learner needs to set a resource limit and choose an arm in such a way that the chance to realize a high reward within the predefined resource limit is high, while the resource limit itself should be kept as low as possible. We derive the theoretical lower bound on the cumulative regret and propose a learning algorithm having a regret upper bound that matches the lower bound. In a simulation study, we show that our learning algorithm outperforms straightforward extensions of standard multi-armed bandit algorithms.

Despite continuous investments in data technologies, the latency of querying data still poses a significant challenge. Modern analytic solutions require near real-time responsiveness both to make them interactive and to support automated processing. Current technologies (Hadoop, Spark, Dataflow) scan the dataset to execute queries. They focus on providing a scalable data storage to maximize task execution speed. We argue that these solutions fail to offer an adequate level of interactivity since they depend on continual access to data. In this paper we present a method for query approximation, also known as approximate query processing (AQP), that reduce the need to scan data during inference (query calculation), thus enabling a rapid query processing tool. We use LSTM network to learn the relationship between queries and their results, and to provide a rapid inference layer for predicting query results. Our method (referred as ``Hunch``) produces a lightweight LSTM network which provides a high query throughput. We evaluated our method using 12 datasets. The results show that our method predicted queries' results with a normalized root mean squared error (NRMSE) ranging from approximately 1\% to 4\%. Moreover, our method was able to predict up to 120,000 queries in a second (streamed together), and with a single query latency of no more than 2ms.

Algorithmic decisions are now being used on a daily basis, and based on Machine Learning (ML) processes that may be complex and biased. This raises several concerns given the critical impact that biased decisions may have on individuals or on society as a whole. Not only unfair outcomes affect human rights, they also undermine public trust in ML and AI. In this paper we address fairness issues of ML models based on decision outcomes, and we show how the simple idea of "feature dropout" followed by an "ensemble approach" can improve model fairness. To illustrate, we will revisit the case of "LimeOut" that was proposed to tackle "process fairness", which measures a model's reliance on sensitive or discriminatory features. Given a classifier, a dataset and a set of sensitive features, LimeOut first assesses whether the classifier is fair by checking its reliance on sensitive features using "Lime explanations". If deemed unfair, LimeOut then applies feature dropout to obtain a pool of classifiers. These are then combined into an ensemble classifier that was empirically shown to be less dependent on sensitive features without compromising the classifier's accuracy. We present different experiments on multiple datasets and several state of the art classifiers, which show that LimeOut's classifiers improve (or at least maintain) not only process fairness but also other fairness metrics such as individual and group fairness, equal opportunity, and demographic parity.

Social norms---the unspoken commonsense rules about acceptable social behavior---are crucial in understanding the underlying causes and intents of people's actions in narratives. For example, underlying an action such as "wanting to call cops on my neighbors" are social norms that inform our conduct, such as "It is expected that you report crimes." We present Social Chemistry, a new conceptual formalism to study people's everyday social norms and moral judgments over a rich spectrum of real life situations described in natural language. We introduce Social-Chem-101, a large-scale corpus that catalogs 292k rules-of-thumb such as "it is rude to run a blender at 5am" as the basic conceptual units. Each rule-of-thumb is further broken down with 12 different dimensions of people's judgments, including social judgments of good and bad, moral foundations, expected cultural pressure, and assumed legality, which together amount to over 4.5 million annotations of categorical labels and free-text descriptions. Comprehensive empirical results based on state-of-the-art neural models demonstrate that computational modeling of social norms is a promising research direction. Our model framework, Neural Norm Transformer, learns and generalizes Social-Chem-101 to successfully reason about previously unseen situations, generating relevant (and potentially novel) attribute-aware social rules-of-thumb.

Public trust in artificial intelligence becomes increasingly crucial as machine-learning companies move from the conceptual to the commercial stage, she …

Kay Firth-Butterfield is Head of AI & ML at World Economic Forum, and a humanitarian with a strong sense of social justice. Kay talks to us about why AI Ethics matter during her presentation at the RE•WORK Applied AI Virtual Summit. Read the full transcript below and watch the video here. It's really great to be with you, and thanks to RE.WORK for making it happen. My title is, Does AI Ethics Matter?

Canada is an engineering powerhouse. Beyond designing, building and maintaining our vital infrastructure, engineering industries form the backbone of the country's economy. Given its power in shaping our environmental, virtual and social landscapes, it's worth asking how the profession can evolve to meet the current moment of compounded health and environmental crises. Like medicine and law, engineering is a regulated profession in Canada. Each province or territory has an official organization responsible for granting P.Eng. These requirements are in place to ensure the safety of both workers and the public when an engineering project is executed, but also to ensure that there is a clear line of accountability if things go wrong.

Written by Camylle Lanteigne (@CamLante), who's currently pursuing a Master's in Public Policy at Concordia University and whose work on social robots and empathy has been featured on Vox. This explainer was written in response to colleagues' requests to know more about temporal bias in AI ethics. It begins with a refresher on cognitive biases, then dives into: how humans understand time, time preferences, present-day preference, confidence changes, planning fallacies, and hindsight bias. Bias is a really big topic, but I'll try to succinctly define a subsection of it--implicit cognitive bias--in a way that is useful for AI ethics, particularly. Humans have cognitive biases, which means every one of us, to varying degrees, holds beliefs and impressions that are not backed up by fleshed out reasoning or evidence, or that we never bothered questioning in the first place.¹

Back in 2013, Microsoft came up with the exciting idea of introducing the general public to Artificial Intelligence. Tay, the teenage chatbot, was launched into the Twittersphere to interact with the platform's audience. Being a young and hip chatbot, it was programmed to use modern slang language instead of formal English. Tay was to mimic those that interacted with her so she could learn the human ways. But this experiment didn't turn out the way Microsoft expected it would. Trolls on the social media site took advantage of Tay's "repeat after me" function and turned her into one of the most bigoted profiles on the forum.