Tesla's "autonomy day" kicked off on Monday morning at the electric-vehicle maker's headquarters in Palo Alto, California, where executives including CEO Elon Musk were expected to give investors more details about the company's self-driving technology, known as Autopilot. "Tesla is making significant progress in the development of its autonomous driving software and hardware, including our FSD computer, which is currently in production and which will enable full-self driving via future over-the-air software updates," the company said when it announced the event. Attendees were given red, Tesla-branded badges with sequential numbers, assumably for test rides of the full self-driving functionality. Musk took the stage just before noon alongside Pete Bannon, the vice president of Autopilot engineering, as more than 40,000 people watched remotely via the company's live YouTube stream. Bannon explained how Tesla designed a new chip for its Autopilot software, noting that the company was able to leverage expertise from multiple teams across the business.

Much of the discussion of nation-state competition in artificial intelligence (AI) focuses on relatively easily quantifiable phenomena including funding, technological advances, access to data and computational power, and the speed of AI industrialization. However, a central element of AI leadership is something much less tangible: control over the norms and values that shape the development and use of AI around the world. The U.S. government has overlooked this dimension of AI development for years, but the last couple months indicate the beginnings of a change of course. If the U.S. hopes to maintain global AI leadership, the government must continue to stake out a comprehensive positive vision, or we may find that the future of AI is a world few of us want to live in. Until recent months, the U.S. government had remained relatively quiet on the topics of AI values and ethics.

From Musk's vantage point, Tesla has a huge advantage over autonomous vehicle competitors because it gathers a massive amount of data in the real world. This quarter, he said Tesla will have 500,000 vehicles on the road, each equipped with eight cameras, ultrasonic sensors and radar gathering data to help build the company's neural network, which will serve as the digital equivalent of the self-driving cars' consciousness.

Tesla, under pressure to show it can generate profits on its main business of making electric cars, on Monday trumpeted a custom-designed computer chip to let its vehicles drive themselves. Even with the new chip -- which comes with all new vehicles and can be installed in older ones -- Teslas still aren't yet fully capable of driving without human intervention. They now have "all hardware necessary," said Elon Musk, Tesla's chief executive officer. "All you have to do is improve the software." The software will be updated over the air to allow full self-driving by the end of the year, he said.

A reliable and accurate forecasting method for crop yields is very important for the farmer, the economy of a country, and the agricultural stakeholders. However, due to weather extremes and uncertainties as a result of increasing climate change, most crop yield forecasting models are not reliable and accurate. In this paper, a hybrid crop yield probability density forecasting method via quantile regression forest and Epanechnikov kernel function (QRF-SJ) is proposed to capture the uncertainties and extremes of weather in crop yield forecasting. By assigning probability to possible crop yield values, probability density forecast gives a complete description of the yield of crops. A case study using the annual crop yield of groundnut and millet in Ghana is presented to illustrate the efficiency and robustness of the proposed technique. The proposed model is able to capture the nonlinearity between crop yield and the weather variables via random forest. The values of prediction interval coverage probability and prediction interval normalized average width for the two crops show that the constructed prediction intervals cover the target values with perfect probability. The probability density curves show that QRF-SJ method has a very high ability to forecast quality prediction intervals with a higher coverage probability. The feature importance gave a score of the importance of each weather variable in building the quantile regression forest model. The farmer and other stakeholders are able to realize the specific weather variable that affect the yield of a selected crop through feature importance. The proposed method and its application on crop yield dataset is the first of its kind in literature.

Model-free learning for multi-agent stochastic games is an active area of research. Existing reinforcement learning algorithms, however, are often restricted to zero-sum games, and are applicable only in small state-action spaces or other simplified settings. Here, we develop a new data efficient Deep-Q-learning methodology for model-free learning of Nash equilibria for general-sum stochastic games. The algorithm uses a local linear-quadratic expansion of the stochastic game, which leads to analytically solvable optimal actions. The expansion is parametrized by deep neural networks to give it sufficient flexibility to learn the environment without the need to experience all state-action pairs. We study symmetry properties of the algorithm stemming from label-invariant stochastic games and as a proof of concept, apply our algorithm to learning optimal trading strategies in competitive electronic markets.

Sparse PCA has shown its effectiveness in high dimensional data analysis, while there is still a gap between the computational method and statistical theory. This paper presents algorithms to solve the row-sparsity constrained PCA, named Feature Sparse PCA (FSPCA), which performs feature selection and PCA simultaneously. Existing techniques to solve the FSPCA problem suffer two main drawbacks: (1) most approaches only solve the leading eigenvector and rely on the deflation technique to estimate the leading m eigenspace, which has feature sparsity inconsistence, identifiability, and orthogonality issues; (2) some approaches are heuristics without convergence guarantee. In this paper, we present convergence guaranteed algorithms to directly estimate the leading m eigenspace. In detail, we show for a low rank covariance matrix, the FSPCA problem can be solved globally (Algorithm 1). Then, we propose an algorithm (Algorithm 2) to solve the FSPCA for general covariance by iteratively building a carefully designed low rank proxy covariance. Theoretical analysis gives the convergence guarantee. Experimental results show the promising performance of the new algorithms compared with the state-of-the-art method on both synthetic and real-world datasets.

The variational autoencoder (VAE) framework remains a popular option for training unsupervised generative models, especially for discrete data where generative adversarial networks (GANs) require workaround to create gradient for the generator. In our work modeling US postal addresses, we show that our discrete VAE with tree recursive architecture demonstrates limited capability of capturing field correlations within structured data, even after overcoming the challenge of posterior collapse with scheduled sampling and tuning of the KL-divergence weight $\beta$. Worse, VAE seems to have difficulty mapping its generated samples to the latent space, as their VAE loss lags behind or even increases during the training process. Motivated by this observation, we show that augmenting training data with generated variants (augmented training) and training a VAE with multiple values of $\beta$ simultaneously (multiscale VAE) both improve the generation quality of VAE. Despite their differences in motivation and emphasis, we show that augmented training and multiscale VAE are actually connected and have similar effects on the model.

In this paper we present a set of key demarcations, particularly important when discussing ethical and societal issues of current AI research and applications. Properly distinguishing issues and concerns related to Artificial General Intelligence and weak AI, between symbolic and connectionist AI, AI methods, data and applications are prerequisites for an informed debate. Such demarcations would not only facilitate much-needed discussions on ethics on current AI technologies and research. In addition sufficiently establishing such demarcations would also enhance knowledge-sharing and support rigor in interdisciplinary research between technical and social sciences.

$\textit{Magic: The Gathering}$ is a popular and famously complicated trading card game about magical combat. In this paper we show that optimal play in real-world $\textit{Magic}$ is at least as hard as the Halting Problem, solving a problem that has been open for a decade. To do this, we present a methodology for embedding an arbitrary Turing machine into a game of $\textit{Magic}$ such that the first player is guaranteed to win the game if and only if the Turing machine halts. Our result applies to how real $\textit{Magic}$ is played, can be achieved using standard-size tournament-legal decks, and does not rely on stochasticity or hidden information. Our result is also highly unusual in that all moves of both players are forced in the construction. This shows that even recognising who will win a game in which neither player has a non-trivial decision to make for the rest of the game is undecidable. We conclude with a discussion of the implications for a unified computational theory of games and remarks about the playability of such a board in a tournament setting.