A tiny California start-up is looking to printers to solve the housing crisis – actually, a very large 3D printer. The company, Mighty Buildings, has been showcasing small (350 square foot) studio apartment models of its new "ADU" units (Accessory Dwelling Units) aimed at backyards and selling for around $115,000. That is, if you do the work and deal with local governments to get all the permits, connect the utilities and install the unit. Have Mighty set it up for you, and you're looking around $184,000. Sam Ruben, the co-founder of the firm, says Mighty can have the home in place in just over two weeks.

Many important real-world settings contain multiple players interacting over an unknown duration with probabilistic state transitions, and are naturally modeled as stochastic games. Prior research on algorithms for stochastic games has focused on two-player zero-sum games, games with perfect information, and games with imperfect-information that is local and does not extend between game states. We present an algorithm for approximating Nash equilibrium in multiplayer general-sum stochastic games with persistent imperfect information that extends throughout game play. We experiment on a 4-player imperfect-information naval strategic planning scenario. Using a new procedure, we are able to demonstrate that our algorithm computes a strategy that closely approximates Nash equilibrium in this game.

Machine learning can be a valuable tool for speeding up elements of the research process. It can be used to analyze data and create knowledge graphs and to surface the most relevant research for a specific research community. But as Dr. Alex Ganose, a postdoctoral researcher at Lawrence Berkeley National Laboratory (LBNL), points out, it needs to be deployed wisely. Alex's research involves using data science and machine learning to solve problems in materials science. "Most recently, I've developed a new framework to calculate electron lifetimes from first principles," he explained.

Directional data consist of observations distributed on a (hyper)sphere, and appear in many applied fields, such as astronomy, ecology, and environmental science. This paper studies both statistical and computational problems of kernel smoothing for directional data. We generalize the classical mean shift algorithm to directional data, which allows us to identify local modes of the directional kernel density estimator (KDE). The statistical convergence rates of the directional KDE and its derivatives are derived, and the problem of mode estimation is examined. We also prove the ascending property of our directional mean shift algorithm and investigate a general problem of gradient ascent on the unit hypersphere. To demonstrate the applicability of our proposed algorithm, we evaluate it as a mode clustering method on both simulated and real-world datasets.

Modeling complex spatial and temporal correlations in the correlated time series data is indispensable for understanding the traffic dynamics and predicting the future status of an evolving traffic system. Recent works focus on designing complicated graph neural network architectures to capture shared patterns with the help of pre-defined graphs. In this paper, we argue that learning node-specific patterns is essential for traffic forecasting while the pre-defined graph is avoidable. To this end, we propose two adaptive modules for enhancing Graph Convolutional Network (GCN) with new capabilities: 1) a Node Adaptive Parameter Learning (NAPL) module to capture node-specific patterns; 2) a Data Adaptive Graph Generation (DAGG) module to infer the inter-dependencies among different traffic series automatically. We further propose an Adaptive Graph Convolutional Recurrent Network (AGCRN) to capture fine-grained spatial and temporal correlations in traffic series automatically based on the two modules and recurrent networks. Our experiments on two real-world traffic datasets show AGCRN outperforms state-of-the-art by a significant margin without pre-defined graphs about spatial connections.

Recent research in Visual Question Answering (VQA) has revealed state-of-the-art models to be inconsistent in their understanding of the world -- they answer seemingly difficult questions requiring reasoning correctly but get simpler associated sub-questions wrong. These sub-questions pertain to lower level visual concepts in the image that models ideally should understand to be able to answer the higher level question correctly. To address this, we first present a gradient-based interpretability approach to determine the questions most strongly correlated with the reasoning question on an image, and use this to evaluate VQA models on their ability to identify the relevant sub-questions needed to answer a reasoning question. Next, we propose a contrastive gradient learning based approach called Sub-question Oriented Tuning (SOrT) which encourages models to rank relevant sub-questions higher than irrelevant questions for an <$image, reasoning-question$> pair. We show that SOrT improves model consistency by upto 6.5% points over existing baselines, while also improving visual grounding.

In this work, we develop DeepWiPHY, a deep learning-based architecture to replace the channel estimation, common phase error (CPE) correction, sampling rate offset (SRO) correction, and equalization modules of IEEE 802.11ax based orthogonal frequency division multiplexing (OFDM) receivers. We first train DeepWiPHY with a synthetic dataset, which is generated using representative indoor channel models and includes typical radio frequency (RF) impairments that are the source of nonlinearity in wireless systems. To further train and evaluate DeepWiPHY with real-world data, we develop a passive sniffing-based data collection testbed composed of Universal Software Radio Peripherals (USRPs) and commercially available IEEE 802.11ax products. The comprehensive evaluation of DeepWiPHY with synthetic and real-world datasets (110 million synthetic OFDM symbols and 14 million real-world OFDM symbols) confirms that, even without fine-tuning the neural network's architecture parameters, DeepWiPHY achieves comparable performance to or outperforms the conventional WLAN receivers, in terms of both bit error rate (BER) and packet error rate (PER), under a wide range of channel models, signal-to-noise (SNR) levels, and modulation schemes.

Many businesses and industries require accurate forecasts for weekly time series nowadays. The forecasting literature however does not currently provide easy-to-use, automatic, reproducible and accurate approaches dedicated to this task. We propose a forecasting method that can be used as a strong baseline in this domain, leveraging state-of-the-art forecasting techniques, forecast combination, and global modelling. Our approach uses four base forecasting models specifically suitable for forecasting weekly data: a global Recurrent Neural Network model, Theta, Trigonometric Box-Cox ARMA Trend Seasonal (TBATS), and Dynamic Harmonic Regression ARIMA (DHR-ARIMA). Those are then optimally combined using a lasso regression stacking approach. We evaluate the performance of our method against a set of state-of-the-art weekly forecasting models on six datasets. Across four evaluation metrics, we show that our method consistently outperforms the benchmark methods by a considerable margin with statistical significance. In particular, our model can produce the most accurate forecasts, in terms of mean sMAPE, for the M4 weekly dataset.

With Covid-19 cases surging in parts of the U.S. at the start of flu season, developers of artificial intelligence tools are about to face their biggest test of the pandemic: Can they help doctors differentiate between the two respiratory illnesses, and accurately predict which patients will become severely ill? Numerous AI models are promising to do exactly that by sifting data on symptoms and analyzing chest X-rays and CT scans. For now, the increased availability of coronavirus testing means AI is unlikely to be relied upon for frontline detection and diagnosis. But it will become increasingly important for figuring out how aggressively to treat patients and which ones are likely to need intensive care beds, ventilators, and other equipment that could become scarce if there's a Covid-flu "twindemic." "That's on the forefront of everyone's mind right now," said Anna Yaffee, an emergency medicine physician at Emory University who helped build an online symptom checker to assess Covid-19 patients.

HitchBot, a friendly-looking talking robot with a bucket for a body and pool-noodle limbs, first arrived on American soil back in 2015. This "hitchhiking" robot was an experiment by a pair of Canadian researchers who wanted to investigate people's trust in, and attitude towards, technology. The researchers wanted to see "whether a robot could hitchhike across the country, relying only on the goodwill and help of strangers." With rudimentary computer vision and a limited vocabulary but no independent means of locomotion, HitchBot was fully dependent on the participation of willing passers-by to get from place to place. Fresh off its successful journey across Canada, where it also picked up a fervent social media following, HitchBot was dropped off in Massachusetts and struck out towards California. But HitchBot never made it to the Golden State.