At a conference at the Vatican last week, Pope Francis warned a group of Silicon Valley execs that in the wrong hands, artificial intelligence could have devastating consequences for humanity. "If mankind's so-called technological progress were to become an enemy of the common good, this would lead to an unfortunate regression to a form of barbarism dictated by the law of the strongest," he said, according to Reuters. The development of advanced AI can "raise increasingly significant implications in all areas of human activity," the Pope said. He also called for "open and concrete discussions" to develop "both theoretical and practical moral principles." The conference also grappled with the March 2019 attacks in Christchurch, New Zealand, and how social media platforms helped spread footage taken during the shootings, according to TIME.

Robert Epstein was looking for love. The year being 2006, he was looking online. As he recounted in the journal Scientific American Mind, he began a promising email exchange with a pretty brunette in Russia. Epstein was disappointed - he wanted more than a penfriend, let's be frank - but she was warm and friendly. Soon she confessed she was developing a crush on him.

A ground-breaking study has revealed that members of the great apes, such as bonobos, chimps and orangutans, have a theory of mind. This, researchers say, proves they can understand others' mental states -- an ability previously though exclusively reserved to humans. The idea other animals possess this trait has been debated for decades and researchers at Kyoto University think they have proved its existence. A ground-breaking study has revealed that members of the great apes, such as bonobos, chimps and orangutans, have a theory of mind. This, researchers say, proves they can understand others' mental states Theory of mind is a higher cognitive function which allows individuals to understand others' mental states.

The rapid development of autonomous vehicles (AVs) holds vast potential for transportation systems through improved safety, efficiency, and access to mobility. However, due to numerous technical, political, and human factors challenges, new methodologies are needed to design vehicles and transportation systems for these positive outcomes. This article tackles important technical challenges arising from the partial adoption of autonomy (hence termed mixed autonomy, to involve both AVs and human-driven vehicles): partial control, partial observation, complex multi-vehicle interactions, and the sheer variety of traffic settings represented by real-world networks. To enable the study of the full diversity of traffic settings, we first propose to decompose traffic control tasks into modules, which may be configured and composed to create new control tasks of interest. These modules include salient aspects of traffic control tasks: networks, actors, control laws, metrics, initialization, and additional dynamics. Second, we study the potential of model-free deep Reinforcement Learning (RL) methods to address the complexity of traffic dynamics. The resulting modular learning framework is called Flow. Using Flow, we create and study a variety of mixed-autonomy settings, including single-lane, multi-lane, and intersection traffic. In all cases, the learned control law exceeds human driving performance (measured by system-level velocity) by at least 40% with only 5-10% adoption of AVs. In the case of partially-observed single-lane traffic, we show that a low-parameter neural network control law can eliminate commonly observed stop-and-go traffic. In particular, the control laws surpass all known model-based controllers, achieving near-optimal performance across a wide spectrum of vehicle densities (even with a memoryless control law) and generalizing to out-of-distribution vehicle densities.

This paper presents the application of an iterative learning control scheme to improve the position tracking performance for an articulated soft robotic arm during aggressive maneuvers. Two antagonistically arranged, inflatable bellows actuate the robotic arm and provide high compliance while enabling fast actuation. Switching valves are used for pressure control of the soft actuators. A norm-optimal iterative learning control scheme based on a linear model of the system is presented and applied in parallel with a feedback controller. The learning scheme is experimentally evaluated on an aggressive trajectory involving set point shifts of 60 degrees within 0.2 seconds. The effectiveness of the learning approach is demonstrated by a reduction of the root-mean-square tracking error from 13 degrees to less than 2 degrees after applying the learning scheme for less than 30 iterations.

We explore the hyperparameter space of reservoir computers used for forecasting of the chaotic Lorenz '63 attractor with Bayesian optimization. We use a new measure of reservoir performance, designed to emphasize learning the global climate of the forecasted system rather than short-term prediction. We find that optimizing over this measure more quickly excludes reservoirs that fail to reproduce the climate. The results of optimization are surprising: the optimized parameters often specify a reservoir network with very low connectivity. Inspired by this observation, we explore reservoir designs with even simpler structure, and find well-performing reservoirs that have zero spectral radius and no recurrence. These simple reservoirs provide counterexamples to widely used heuristics in the field, and may be useful for hardware implementations of reservoir computers.

A BSTRACT We empirically evaluate common assumptions about neural networks that are widely held by practitioners and theorists alike. We study the prevalence of local minima in loss landscapes, whether small-norm parameter vectors generalize better (and whether this explains the advantages of weight decay), whether wide-network theories (like the neural tangent kernel) describe the behaviors of classifiers, and whether the rank of weight matrices can be linked to generalization and robustness in real-world networks. In statistical learning, principled kernel methods have vastly improved the performance of SVMs and PCA (Suykens & V andewalle, 1999; Sch olkopf et al., 1997), and boosting theory has enabled weak learners to generate strong classifiers (Schapire, 1990). Optimizers in deep learning are borrowed from the field of convex optimization, where momentum optimizers (Nesterov, 1983) and conjugate gradient methods provably solve ill-conditioned problems with high efficiency (Hestenes & Stiefel, 1952). Deep learning harnesses foundational tools from these mature parent fields. Despite its rigorous roots, deep learning has driven a wedge between theory and practice. Recent theoretical work has certainly made impressive strides towards understanding optimization and generalization in neural networks. But doing so has required researchers to make strong assumptions and study restricted model classes. In this paper, we seek to understand whether deep learning theories accurately capture the behaviors and network properties that make realistic deep networks work. Following a line of previous work, such as Swirszcz et al. (2016), Zhang et al. (2016), Balduzzi et al. (2017) and Santurkar et al. (2018), we put the assumptions and conclusions of deep learning theory to the test using experiments with both toy networks and realistic ones. We focus on the following important theoretical issues: - Local minima: Numerous theoretical works argue that all local minima of neural loss functions are globally optimal or that all local minima are nearly optimal. In practice, we find Authors contributed equally. 1 arXiv:1910.00359v1 Y et for neural networks, it is not at all clear which form of null 2-regularization is optimal.

Molecule generation is to design new molecules with specific chemical properties and further to optimize the desired chemical properties. Following previous work, we encode molecules into continuous vectors in the latent space and then decode the vectors into molecules under the variational autoencoder (VAE) framework. We investigate the posterior collapse problem of current RNN-based VAEs for molecule sequence generation. For the first time, we find that underestimated reconstruction loss leads to posterior collapse, and provide both theoretical and experimental evidence. We propose an effective and efficient solution to fix the problem and avoid posterior collapse. Without bells and whistles, our method achieves SOTA reconstruction accuracy and competitive validity on the ZINC 250K dataset. When generating 10,000 unique valid SMILES from random prior sampling, it costs JT-VAE1450s while our method only needs 9s. Our implementation is at https://github.com/chaoyan1037/Re-balanced-VAE.

This work provides the first unifying theoretical framework for node embeddings and structural graph representations, bridging methods like matrix factorization and graph neural networks. Using invariant theory, we show that the relationship between structural representations and node embeddings is analogous to that of a distribution and its samples. We prove that all tasks that can be performed by node embeddings can also be performed by structural representations and vice-versa. We also show that the concept of transductive and inductive learning is unrelated to node embeddings and graph representations, clearing another source of confusion in the literature. Finally, we introduce new practical guidelines to generating and using node embeddings, which fixes significant shortcomings of standard operating procedures used today.

The goal of lifetime clustering is to develop an inductive model that maps subjects into $K$ clusters according to their underlying (unobserved) lifetime distribution. We introduce a neural-network based lifetime clustering model that can find cluster assignments by directly maximizing the divergence between the empirical lifetime distributions of the clusters. Accordingly, we define a novel clustering loss function over the lifetime distributions (of entire clusters) based on a tight upper bound of the two-sample Kuiper test p-value. The resultant model is robust to the modeling issues associated with the unobservability of termination signals, and does not assume proportional hazards. Our results in real and synthetic datasets show significantly better lifetime clusters (as evaluated by C-index, Brier Score, Logrank score and adjusted Rand index) as compared to competing approaches.