This paper explores how different ideas of racial equity in machine learning, in justice settings in particular, can present trade-offs that are difficult to solve computationally. Machine learning is often used in justice settings to create risk assessments, which are used to determine interventions, resources, and punitive actions. Overall aspects and performance of these machine learning-based tools, such as distributions of scores, outcome rates by levels, and the frequency of false positives and true positives, can be problematic when examined by racial group. Models that produce different distributions of scores or produce a different relationship between level and outcome are problematic when those scores and levels are directly linked to the restriction of individual liberty and to the broader context of racial inequity. While computation can help highlight these aspects, data and computation are unlikely to solve them. This paper explores where values and mission might have to fill the spaces computation leaves.

Moderate-size quantum computers are now publicly accessible over the cloud, opening the exciting possibility of performing dynamical simulations of quantum systems. However, while rapidly improving, these devices have short coherence times, limiting the depth of algorithms that may be successfully implemented. Here we demonstrate that, despite these limitations, it is possible to implement long-time, high fidelity simulations on current hardware. Specifically, we simulate an XY-model spin chain on the Rigetti and IBM quantum computers, maintaining a fidelity of at least 0.9 for over 600 time steps. This is a factor of 150 longer than is possible using the iterated Trotter method. Our simulations are performed using a new algorithm that we call the fixed state Variational Fast Forwarding (fsVFF) algorithm. This algorithm decreases the circuit depth and width required for a quantum simulation by finding an approximate diagonalization of a short time evolution unitary. Crucially, fsVFF only requires finding a diagonalization on the subspace spanned by the initial state, rather than on the total Hilbert space as with previous methods, substantially reducing the required resources.

Cyber-defense systems are being developed to automatically ingest Cyber Threat Intelligence (CTI) that contains semi-structured data and/or text to populate knowledge graphs. A potential risk is that fake CTI can be generated and spread through Open-Source Intelligence (OSINT) communities or on the Web to effect a data poisoning attack on these systems. Adversaries can use fake CTI examples as training input to subvert cyber defense systems, forcing the model to learn incorrect inputs to serve their malicious needs. In this paper, we automatically generate fake CTI text descriptions using transformers. We show that given an initial prompt sentence, a public language model like GPT-2 with fine-tuning, can generate plausible CTI text with the ability of corrupting cyber-defense systems. We utilize the generated fake CTI text to perform a data poisoning attack on a Cybersecurity Knowledge Graph (CKG) and a cybersecurity corpus. The poisoning attack introduced adverse impacts such as returning incorrect reasoning outputs, representation poisoning, and corruption of other dependent AI-based cyber defense systems. We evaluate with traditional approaches and conduct a human evaluation study with cybersecurity professionals and threat hunters. Based on the study, professional threat hunters were equally likely to consider our fake generated CTI as true.

Model Agnostic Meta-Learning (MAML) has emerged as a standard framework for meta-learning, where a meta-model is learned with the ability of fast adapting to new tasks. However, as a double-looped optimization problem, MAML needs to differentiate through the whole inner-loop optimization path for every outer-loop training step, which may lead to both computational inefficiency and sub-optimal solutions. In this paper, we generalize MAML to allow meta-learning to be defined in function spaces, and propose the first meta-learning paradigm in the Reproducing Kernel Hilbert Space (RKHS) induced by the meta-model's Neural Tangent Kernel (NTK). Within this paradigm, we introduce two meta-learning algorithms in the RKHS, which no longer need a sub-optimal iterative inner-loop adaptation as in the MAML framework. We achieve this goal by 1) replacing the adaptation with a fast-adaptive regularizer in the RKHS; and 2) solving the adaptation analytically based on the NTK theory. Extensive experimental studies demonstrate advantages of our paradigm in both efficiency and quality of solutions compared to related meta-learning algorithms. Another interesting feature of our proposed methods is that they are demonstrated to be more robust to adversarial attacks and out-ofdistribution adaptation than popular baselines, as demonstrated in our experiments. Meta-learning (Schmidhuber, 1987) has made tremendous progresses in the last few years. It aims to learn abstract knowledge from many related tasks so that fast adaption to new and unseen tasks becomes possible. For example, in few-shot learning, meta-learning corresponds to learning a meta-model or meta-parameters so that they can fast adapt to new tasks with a limited number of data samples.

The internet today is nothing like the World Wide Web that Sir Tim Berners-Lee envisioned back when he invented it in 1989. While it continues to be a place where people can interact in a free exchange of ideas, individuals and groups have lost their sense of empowerment to a handful of giant monopolies and countries that are bent on collecting their personal data. "Once a platform becomes dominant, it is able to collect more data," said Pieter Verdegem, senior lecturer at the University of Westminster School of Media and Communication. "That is what we are seeing around the world, and it explains why we have the so-called GAFAM – Google, Apple, Facebook, Amazon and Microsoft – in the U.S. and the so-called BAT – Baidu, Alibaba and Tencent – in China." "The next wave is about artificial intelligence," Verdegem said. "Companies and governments will use all that data to train algorithms to come up with better deep learning models.

Artificial intelligence (AI) is an ever-evolving technology. With several different uses, it's easy to understand why it's being implemented more and more frequently. These titles answer common questions about AI, discuss what current AI technologies businesses are using, how humans can lose control over AI, and more. In T-Minus AI, author, national expert, and the US Air Force's first Chairperson for Artificial Intelligence Michael Kanaan explains a human-oriented perspective of AI. He offers his view on our history of innovation to illustrate what we should all know about modern computing, AI, and machine learning.

AI and machine learning may still be hot and top of mind for technology decision makers, line of business folk and investors, but that didn't prevent 2020 from broadsiding some AI initiatives. In our Vote AI ML Infrastructure 2020 survey published in July, 58% of organizations surveyed expected COVID-19 to have a negative impact on their existing AI initiatives, and 19% said the pandemic has led them to stop work on these projects. But 75% of organizations said COVID-19 led to new AI initiatives. In our just-published data from Vote AI ML Use Cases 2021 survey, the picture has changed and things look more optimistic, with 86% of respondents agreeing that the pandemic has or will cause their organization to invest in new AI initiatives. With pandemic-induced uncertainty still looming over us all, this report looks at some of the key trends in AI we expect to see in 2021.

The paper describes a Multisource AI Scorecard Table (MAST) that provides the developer and user of an artificial intelligence (AI)/machine learning (ML) system with a standard checklist focused on the principles of good analysis adopted by the intelligence community (IC) to help promote the development of more understandable systems and engender trust in AI outputs. Such a scorecard enables a transparent, consistent, and meaningful understanding of AI tools applied for commercial and government use. A standard is built on compliance and agreement through policy, which requires buy-in from the stakeholders. While consistency for testing might only exist across a standard data set, the community requires discussion on verification and validation approaches which can lead to interpretability, explainability, and proper use. The paper explores how the analytic tradecraft standards outlined in Intelligence Community Directive (ICD) 203 can provide a framework for assessing the performance of an AI system supporting various operational needs. These include sourcing, uncertainty, consistency, accuracy, and visualization. Three use cases are presented as notional examples that support security for comparative analysis.

Gaussian processes (GPs) are ubiquitously used in sciences and engineering as metamodels. Standard GPs, however, can only handle numerical or quantitative variables. In this paper, we introduce latent map Gaussian processes (LMGPs) that inherit the attractive properties of GPs but are also applicable to mixed data that have both quantitative and qualitative inputs. The core idea behind LMGPs is to learn a low-dimensional manifold where all qualitative inputs are represented by some quantitative features. To learn this manifold, we first assign a unique prior vector representation to each combination of qualitative inputs. We then use a linear map to project these priors on a manifold that characterizes the posterior representations. As the posteriors are quantitative, they can be straightforwardly used in any standard correlation function such as the Gaussian. Hence, the optimal map and the corresponding manifold can be efficiently learned by maximizing the Gaussian likelihood function. Through a wide range of analytical and real-world examples, we demonstrate the advantages of LMGPs over state-of-the-art methods in terms of accuracy and versatility. In particular, we show that LMGPs can handle variable-length inputs and provide insights into how qualitative inputs affect the response or interact with each other. We also provide a neural network interpretation of LMGPs and study the effect of prior latent representations on their performance.

State-of-the-art deep-learning systems use decision rules that are challenging for humans to model. Explainable AI (XAI) attempts to improve human understanding but rarely accounts for how people typically reason about unfamiliar agents. We propose explicitly modeling the human explainee via Bayesian Teaching, which evaluates explanations by how much they shift explainees' inferences toward a desired goal. We assess Bayesian Teaching in a binary image classification task across a variety of contexts. Absent intervention, participants predict that the AI's classifications will match their own, but explanations generated by Bayesian Teaching improve their ability to predict the AI's judgements by moving them away from this prior belief. Bayesian Teaching further allows each case to be broken down into sub-examples (here saliency maps). These sub-examples complement whole examples by improving error detection for familiar categories, whereas whole examples help predict correct AI judgements of unfamiliar cases.