The White House Office of Science and Technology Policy's bill of rights for an artificial intelligence-powered world needs "teeth," in the form of procurement enforcement, said Director Eric Lander on Tuesday. Many AI ethics proposals are little more than a set of basic expectations around governance, privacy, fairness, transparency and explainability, when laws and litigation are needed to back them up, Lander said, during Stanford University's Human-Centered AI Fall Conference. Lander's comments come after the Office of Science and Technology Policy (OSTP) issued a request for information last month on biometrics use cases -- given the technologies' wide adoption for identification, surveillance and behavioral analysis -- to inform development of the AI bill of rights. "We see this as a way not to limit innovation," Lander said. "We see this [as a way] to improve the quality of products by not rewarding people who cut corners and instead setting ground rules to reward people who produce safe, effective, fair, equitable products."

Our problem was classifying the faces of the leader of the political parties during the 2021 Canadian federal election. The dataset contains 497 photos, which were scared from the web, with an equal distribution between the five classes. Below is an example from each of the five classes. The possible classes for the classification model are Annamie Paul, Jagmeet Singh, Justin Trudeau, Erin O'Tool, and Yves-Francois Blanchet. The data is split into training and testing datasets containing 431 images and 66 images, respectfully.

The emergence of Artificial Intelligence (AI) has captured the attention of nearly every industry -- including the public sector. In fact, this June, Lynne Parker became the United States' first AI czar, tasked with evaluating societal risks associated with AI and preventing harm from new technologies. More recently, the U.S. Department of Commerce announced plans to form a committee to advise federal agencies on AI research and development, and the National Artificial Intelligence Advisory committee said they plan to focus on several issues -- including U.S. competitiveness and how AI can enhance opportunities for geographic regions. What's become increasingly clear today -- supported by the continued investment of AI spending -- is that the administration is adamant about bolstering AI efforts and addressing AI regulations across the public and private sectors. Yet, many still wonder whether the increased focus on AI will result in any tangible outcomes.

"Unlike other exoplanet-detecting machine learning programs, ExoMiner isn't a black box – there is no mystery as to why it decides something is a planet or not," said Jon Jenkins, exoplanet scientist at NASA's Ames Research Center in California's Silicon Valley. "We can easily explain which features in the data lead ExoMiner to reject or confirm a planet." What is the difference between a confirmed and validated exoplanet? A planet is "confirmed," when different observation techniques reveal features that can only be explained by a planet. A planet is "validated" using statistics – meaning how likely or unlikely it is to be a planet based on the data.

The Food and Drug Administration recently sought comments on the role of transparency for artificial intelligence AI and machine learning-enabled …

We address a three-tier numerical framework based on manifold learning for the forecasting of high-dimensional time series. At the first step, we embed the time series into a reduced low-dimensional space using a nonlinear manifold learning algorithm such as Locally Linear Embedding and Diffusion Maps. At the second step, we construct reduced-order regression models on the manifold, in particular Multivariate Autoregressive (MVAR) and Gaussian Process Regression (GPR) models, to forecast the embedded dynamics. At the final step, we lift the embedded time series back to the original high-dimensional space using Radial Basis Functions interpolation and Geometric Harmonics. For our illustrations, we test the forecasting performance of the proposed numerical scheme with four sets of time series: three synthetic stochastic ones resembling EEG signals produced from linear and nonlinear stochastic models with different model orders, and one real-world data set containing daily time series of 10 key foreign exchange rates (FOREX) spanning the time period 03/09/2001-29/10/2020. The forecasting performance of the proposed numerical scheme is assessed using the combinations of manifold learning, modelling and lifting approaches. We also provide a comparison with the Principal Component Analysis algorithm as well as with the naive random walk model and the MVAR and GPR models trained and implemented directly in the high-dimensional space.

Cyber attacks are increasing in volume, frequency, and complexity. In response, the security community is looking toward fully automating cyber defense systems using machine learning. However, so far the resultant effects on the coevolutionary dynamics of attackers and defenders have not been examined. In this whitepaper, we hypothesise that increased automation on both sides will accelerate the coevolutionary cycle, thus begging the question of whether there are any resultant fixed points, and how they are characterised. Working within the threat model of Locked Shields, Europe's largest cyberdefense exercise, we study blackbox adversarial attacks on network classifiers. Given already existing attack capabilities, we question the utility of optimal evasion attack frameworks based on minimal evasion distances. Instead, we suggest a novel reinforcement learning setting that can be used to efficiently generate arbitrary adversarial perturbations. We then argue that attacker-defender fixed points are themselves general-sum games with complex phase transitions, and introduce a temporally extended multi-agent reinforcement learning framework in which the resultant dynamics can be studied. We hypothesise that one plausible fixed point of AI-NIDS may be a scenario where the defense strategy relies heavily on whitelisted feature flow subspaces. Finally, we demonstrate that a continual learning approach is required to study attacker-defender dynamics in temporally extended general-sum games.

Automatic summarization of legal texts is an important and still a challenging task since legal documents are often long and complicated with unusual structures and styles. Recent advances of deep models trained end-to-end with differentiable losses can well-summarize natural text, yet when applied to the legal domain, they show limited results. In this paper, we propose to use reinforcement learning to train current deep summarization models to improve their performance in the legal domain. To this end, we adopt proximal policy optimization methods and introduce novel reward functions that encourage the generation of candidate summaries satisfying both lexical and semantic criteria. We apply our method to training different summarization backbones and observe a consistent and significant performance gain across three public legal datasets.

Several problems in neuroimaging and beyond require inference on the parameters of multi-task sparse hierarchical regression models. Examples include M/EEG inverse problems, neural encoding models for task-based fMRI analyses, and climate science. In these domains, both the model parameters to be inferred and the measurement noise may exhibit a complex spatio-temporal structure. Existing work either neglects the temporal structure or leads to computationally demanding inference schemes. Overcoming these limitations, we devise a novel flexible hierarchical Bayesian framework within which the spatio-temporal dynamics of model parameters and noise are modeled to have Kronecker product covariance structure. Inference in our framework is based on majorization-minimization optimization and has guaranteed convergence properties. Our highly efficient algorithms exploit the intrinsic Riemannian geometry of temporal autocovariance matrices. For stationary dynamics described by Toeplitz matrices, the theory of circulant embeddings is employed. We prove convex bounding properties and derive update rules of the resulting algorithms. On both synthetic and real neural data from M/EEG, we demonstrate that our methods lead to improved performance.

Of all the potentially new and revolutionary technologies, artificial intelligence (AI) may be the most disruptive of all. In layman's terms, AI refers to systems able to perform tasks that normally require human intelligence, such as visual and speech recognition, decisionmaking and, perhaps one day, thinking. AI has already defeated the world's best chess and Pokemon GO players. Suppose AI surpasses the intelligence of human beings. Could AI's super-intelligence cure cancer, enhance wellbeing, redress climate change and deal with many of the planet's worst evils?