Time series forecasting is an important problem across many domains, including predictions of solar plant energy output, electricity consumption, and traffic jam situation. In this paper, we propose to tackle such forecasting problem with Transformer. Although impressed by its performance in our preliminary study, we found its two major weaknesses: (1) locality-agnostics: the point-wise dot-product self attention in canonical Transformer architecture is insensitive to local context, which can make the model prone to anomalies in time series; (2) memory bottleneck: space complexity of canonical Transformer grows quadratically with sequence length $L$, making modeling long time series infeasible. In order to solve these two issues, we first propose convolutional self attention by producing queries and keys with causal convolution so that local context can be better incorporated into attention mechanism. Then, we propose LogSparse Transformer with only $O(L(\log L)^{2})$ memory cost, improving the time series forecasting in finer granularity under constrained memory budget. Our experiments on both synthetic data and real-world datasets show that it compares favorably to the state-of-the-art.

Mobile sequential recommendation was originally designed to find a promising route for a single taxicab. Directly applying it for multiple taxicabs may cause an excessive overlap of recommended routes. The multi-taxicab recommendation problem is challenging and has been less studied. In this paper, we first formalize a collective mobile sequential recommendation problem based on a classic mathematical model, which characterizes time-varying influence among competing taxicabs. Next, we propose a new evaluation metric for a collection of taxicab routes aimed to minimize the sum of potential travel time. We then develop an efficient algorithm to calculate the metric and design a greedy recommendation method to approximate the solution. Finally, numerical experiments show the superiority of our methods. In trace-driven simulation, the set of routes recommended by our method significantly outperforms those obtained by conventional methods.

In the spring of 1964, as fighting escalated in Vietnam, several dozen Americans gathered to play a game. They were some of the most powerful men in Washington: the director of Central Intelligence, the Army chief of staff, the national security advisor, and the head of the Strategic Air Command. Senior officials from the State Department and the Navy were also on hand. Players were divided into two teams, red and blue, representing the Cold War superpowers. The teams operated out of separate rooms in the Pentagon, role-playing confrontation in Southeast Asia, simulated in a neutral command center.

WASHINGTON - The Trump administration has moved ahead with a surveillance drone sale to four U.S. allies in the South China Sea region as acting Defense Secretary Patrick Shanahan said Washington will no longer "tiptoe" around Chinese behavior in Asia. The drones would afford greater intelligence gathering capabilities potentially curbing Chinese activity in the region. Shanahan did not directly name China when making accusations of "actors" destabilizing the region in a speech at the annual Shangri-La Dialogue in Singapore on Saturday but went on to say the United States would not ignore Chinese behavior. The Pentagon announced on Friday it would sell 34 ScanEagle drones, made by Boeing Co., to the governments of Malaysia, Indonesia, the Philippines and Vietnam for a total of $47 million. China claims almost all of the strategic South China Sea and frequently lambastes the United States and its allies over naval operations near Chinese-occupied islands.

The Markov decision process (MDP) formulation used to model many real-world sequential decision making problems does not capture the setting where the set of available decisions (actions) at each time step is stochastic. Recently, the stochastic action set Markov decision process (SAS-MDP) formulation has been proposed, which captures the concept of a stochastic action set. In this paper we argue that existing RL algorithms for SAS-MDPs suffer from divergence issues, and present new algorithms for SAS-MDPs that incorporate variance reduction techniques unique to this setting, and provide conditions for their convergence. We conclude with experiments that demonstrate the practicality of our approaches using several tasks inspired by real-life use cases wherein the action set is stochastic.

Traditional approaches focus on finding relationships between two entire time series, however, many interesting relationships exist in small sub-intervals of time and remain feeble during other sub-intervals. We define the notion of a sub-interval relationship (SIR) to capture such interactions that are prominent only in certain sub-intervals of time. To that end, we propose a fast-optimal guaranteed algorithm to find most interesting SIR relationship in a pair of time series. Lastly, we demonstrate the utility of our method in climate science domain based on a real-world dataset along with its scalability scope and obtain useful domain insights.

On Tuesday, in an 8-1 tally, the San Francisco Board of Supervisors voted to ban the use of facial recognition software by city departments, including police. Supporters of the ban cited racial inequality in audits of facial recognition software from companies like Amazon and Microsoft, as well as dystopian surveillance happening now in China. At the core of arguments around the regulation of facial recognition software use is the question of whether a temporary moratorium should be put in place until police and governments adopt policies and standards or it should be permanently banned. Some believe facial recognition software can be used to exonerate the innocent and that more time is needed to gather information. Others, like San Francisco Supervisor Aaron Peskin, believe that even if AI systems achieve racial parity, facial recognition is a "uniquely dangerous and oppressive technology."

Despite a tremendous amount of work in the literature and in the commercial sectors, current approaches to multi-modal trip planning still fail to consistently generate plans that users deem optimal in practice. We believe that this is due to the fact that current planners fail to capture the true preferences of users, e.g., their preferences depend on aspects that are not modeled. An example of this could be a preference not to walk through an unsafe area at night. We present a novel multi-modal trip planner that allows users to up- load auxiliary geographic data (e.g., crime rates) and to specify temporal constraints and preferences over these data in combination with typical metrics such as time and cost. Concretely, our planner supports the modes walking, biking, driving, public transit, and taxi, uses linear temporal logic to capture temporal constraints, and preferential cost functions to represent preferences. We show by examples that this allows the expression of very interesting preferences and constraints that, naturally, lead to quite diverse optimal plans.

The Canadian city of Toronto is a thriving hub of start-up activity in areas ranging from next-generation marketing to AI, fintech and more. As one of the biggest cities in Canada and the capital of the province of Ontario, Toronto is in many respects Canada's start-up capital. The city's tech scene is booming. According to Toronto Global, the city and its surrounding region generated more tech jobs in the previous year than New York City or the San Francisco Bay Area combined. Google has invested $5m in the Vector Institute to make Toronto one of the foremost global players in the AI space.

Scientists have extracted a 400-year record of El Niño events using coral reef cores drilled from the Pacific Ocean, revealing crucial new insight on how these weather patterns have changed. And, the data so far suggest something'unusual' has been happening in recent decades. According to the new research, El Niño events appear to be cropping up more frequently in the central Pacific than they have in past centuries, and while eastern El Niños may be getting stronger. El Niño is caused by a shift in the distribution of warm water in the Pacific Ocean around the equator. Usually the wind blows strongly from east to west, due to the rotation of the Earth, causing water to pile up in the western part of the Pacific.