Domain shift is a fundamental problem in visual recognition which typically arises when the source and target data follow different distributions. The existing domain adaptation approaches which tackle this problem work in the closed-set setting with the assumption that the source and the target data share exactly the same classes of objects. In this paper, we tackle a more realistic problem of open-set domain shift where the target data contains additional classes that are not present in the source data. More specifically, we introduce an end-to-end Progressive Graph Learning (PGL) framework where a graph neural network with episodic training is integrated to suppress underlying conditional shift and adversarial learning is adopted to close the gap between the source and target distributions. Compared to the existing open-set adaptation approaches, our approach guarantees to achieve a tighter upper bound of the target error. Extensive experiments on three standard open-set benchmarks evidence that our approach significantly outperforms the state-of-the-arts in open-set domain adaptation.

This paper addresses the trade-off between Accuracy and Transparency for deep learning applied to sports analytics. Neural nets achieve great predictive accuracy through deep learning, and are popular in sports analytics. But it is hard to interpret a neural net model and harder still to extract actionable insights from the knowledge implicit in it. Therefore, we built a simple and transparent model that mimics the output of the original deep learning model and represents the learned knowledge in an explicit interpretable way. Our mimic model is a linear model tree, which combines a collection of linear models with a regression-tree structure. The tree version of a neural network achieves high fidelity, explains itself, and produces insights for expert stakeholders such as athletes and coaches. We propose and compare several scalable model tree learning heuristics to address the computational challenge from datasets with millions of data points.

The use of heuristics to assess the convergence and compress the output of Markov chain Monte Carlo can be sub-optimal in terms of the empirical approximations that are produced. Typically a number of the initial states are attributed to "burn in" and removed, whilst the remainder of the chain is "thinned" if compression is also required. In this paper we consider the problem of retrospectively selecting a subset of states, of fixed cardinality, from the sample path such that the approximation provided by their empirical distribution is close to optimal. A novel method is proposed, based on greedy minimisation of a kernel Stein discrepancy, that is suitable for problems where heavy compression is required. Theoretical results guarantee consistency of the method and its effectiveness is demonstrated in the challenging context of parameter inference for ordinary differential equations. Software is available in the Stein Thinning package in both Python and MATLAB.

The broad deployment of 802.11 (a.k.a., WiFi) access points and significant enhancement of the energy efficiency of these wireless transceivers has resulted in increasing interest in building 802.11-based IoT systems. Unfortunately, the main energy efficiency mechanisms of 802.11, namely PSM and APSD, fall short when used in IoT applications. PSM increases latency and intensifies channel access contention after each beacon instance, and APSD does not inform stations about when they need to wake up to receive their downlink packets. In this paper, we present a new mechanism---edge-assisted predictive sleep scheduling (EAPS)---to adjust the sleep duration of stations while they expect downlink packets. We first implement a Linux-based access point that enables us to collect parameters affecting communication latency. Using this access point, we build a testbed that, in addition to offering traffic pattern customization, replicates the characteristics of real-world environments. We then use multiple machine learning algorithms to predict downlink packet delivery. Our empirical evaluations confirm that when using EAPS the energy consumption of IoT stations is as low as PSM, whereas the delay of packet delivery is close to the case where the station is always awake.

Rich user behavior data has been proven to be of great value for click-through rate prediction tasks, especially in industrial applications such as recommender systems and online advertising. Both industry and academy have paid much attention to this topic and propose different approaches to modeling with long sequential user behavior data. Among them, memory network based model MIMN proposed by Alibaba, achieves SOTA with the co-design of both learning algorithm and serving system. MIMN is the first industrial solution that can model sequential user behavior data with length scaling up to 1000. However, MIMN fails to precisely capture user interests given a specific candidate item when the length of user behavior sequence increases further, say, by 10 times or more. This challenge exists widely in previously proposed approaches. In this paper, we tackle this problem by designing a new modeling paradigm, which we name as Search-based Interest Model (SIM). SIM extracts user interests with two cascaded search units: (i) General Search Unit acts as a general search from the raw and arbitrary long sequential behavior data, with query information from candidate item, and gets a Sub user Behavior Sequence which is relevant to candidate item; (ii) Exact Search Unit models the precise relationship between candidate item and SBS. This cascaded search paradigm enables SIM with a better ability to model lifelong sequential behavior data in both scalability and accuracy. Apart from the learning algorithm, we also introduce our hands-on experience on how to implement SIM in large scale industrial systems. Since 2019, SIM has been deployed in the display advertising system in Alibaba, bringing 7.1\% CTR and 4.4\% RPM lift, which is significant to the business. Serving the main traffic in our real system now, SIM models user behavior data with maximum length reaching up to 54000, pushing SOTA to 54x.

Microsoft internal sales executives who manage a large number of accounts operate in a challenging environment. They sell a rich suite of products using an assortment of different sales tools and fragmented data. As a result, they spend too much time gathering and verifying customer information, and too little time helping customers realize how they can achieve their business goals through Microsoft technologies. Microsoft is hardly alone in this. Distilling compelling insights from disparate, siloed information systems has historically been a complex and time-consuming task for sales executives in all industries. A holistic view of data and insights at the commercial-account level simply hasn't been available. For sellers, the challenge is that too many tools take too much of their time away from focusing on their customers.

The AI in Enterprise Accounting Market has witnessed continuous growth in the past few years and is projected to grow even further during the forecast period (2020-2025). The assessment provides a 360 view and insights, outlining the key outcomes of the industry. These insights help the business decision-makers to formulate better business plans and make informed decisions for improved profitability. In addition, the study helps venture or private players in understanding the companies more precisely to make better-informed decisions. The AI in Enterprise Accounting Market study covers current status, % share, future patterns, development rate, SWOT examination, sales channels, to anticipate growth scenarios for years 2020-2025.

Drowning is the third leading cause of accidental death, according to World Health Organization (WHO) data, with an estimated 320,000 fatalities each year globally. While lifeguards play a crucial role in helping safeguard beaches and pools, the human eye struggles to spot swimmers in distress in large crowds or at a distance -- with or without the help of binoculars. Sightbit is harnessing AI to alert lifeguards to potential drowning incidents, as well as flagging other hazardous situations, such as unattended children and rip currents. Founded in 2019, Israel-based Sightbit is a spinout from Ben-Gurion University of the Negev (BGU). The public research university invests in alumni via its Cactus Capital VC fund and has provided pre-seed funding to Sightbit, which is currently raising additional funds as part of a seed round.

Oceans are the driving force of Mother Nature, holding 97% of earth's water. Oceanic ecosystems involve many critical marine species such as fishes, seagrasses, and coral reefs. These are essential in the marine ecosystem, for example, if seagrasses are removed, this may lead to the reduction of light required for photosynthesis. At the same time, it involves huge maintenance of these marine species. Due to tourism, shipping, and human intervention, 75% of the world's coral reefs are being threatened and 19% of the coral reefs having been destroyed by 2011.

In times of crisis, we all want to know where the robots are! And young roboticists just starting their careers, or simply thinking about robotics as a career, ask us'How can robotics help?' and'What can I do to help?'. Cluster organizations like Silicon Valley Robotics can serve as connection points between industry and academia, between undergrads and experts, between startups and investors, which is why we rapidly organized a weekly discussion with experts about "COVID-19, robots and us" (video playlist). During our online series, we heard from roboticists directly helping with all sorts of COVID-19 response, like Gui Cavalcanti of Open Source Medical Supplies and Alder Riley of Helpful Engineering. Both groups are great examples of the incredible power of people working together.