Sampling from an unnormalized target distribution is an essential problem with many applications in probabilistic inference. Stein Variational Gradient Descent (SVGD) has been shown to be a powerful method that iteratively updates a set of particles to approximate the distribution of interest. Furthermore, when analysing its asymptotic properties, SVGD reduces exactly to a single-objective optimization problem and can be viewed as a probabilistic version of this single-objective optimization problem. A natural question then arises: ``Can we derive a probabilistic version of the multi-objective optimization?''. To answer this question, we propose Stochastic Multiple Target Sampling Gradient Descent (MT-SGD), enabling us to sample from multiple unnormalized target distributions. Specifically, our MT-SGD conducts a flow of intermediate distributions gradually orienting to multiple target distributions, which allows the sampled particles to move to the joint high-likelihood region of the target distributions. Interestingly, the asymptotic analysis shows that our approach reduces exactly to the multiple-gradient descent algorithm for multi-objective optimization, as expected. Finally, we conduct comprehensive experiments to demonstrate the merit of our approach to multi-task learning.

Around 30% of undergraduate college and university students in the United States fall under the category of first-generation and/or low-income (FLI) [1].Within science, technology, engineering, and mathematics (STEM), the number of FLI students is closer to 20% [2], indicating a disparity in support and accessibility for FLI students in STEM.FLI students are an important part of the American higher educational landscape yet much of the rhetoric and research on FLI college students is centered on deficit-based frameworks [3] that focus on what they lack, rather than leveraging their strengths.FLI students enrich educational spaces-especially in STEM environments-and should be met with proper support so that they may thrive in their programs.Research shows that FLI students offer robust life experiences and are innovative, collaborative, community-oriented, and creative [3-8].Educators must be ready to support them through asset-based approaches to improve FLI STEM student retention.This paper offers 10 simple rules for successfully supporting FLI STEM undergraduate students.The strategies below are informed by research as well as the authors' experiences as education researchers, practitioners, and former FLI students.It will be most helpful for mentors, advisors, professors, or anyone in a supportive role for FLI undergraduate students in STEM fields. Rule 1: Get to know the FLI community on campusFLI students exist on every campus, yet because this population is centered around socioeconomic status, a student's FLI identity might not always be apparent or visible, making it difficult to know when to differentiate support.A great first step in supporting FLI students is to get to know what exists and build from there [9].For example, many campuses have university-sponsored programming and resources, as well as student-led organizations created specifically to support FLI students.You can begin by searching your institution's website for FLI student resources or by asking colleagues.It might take some work to get to know what organizations and resources exist to support FLI students at your institution, but there are many benefits to being engaged with this community.

Max-linear Bayesian networks have emerged as highly applicable models for causal inference from extreme value data. However, conditional independence (CI) for max-linear Bayesian networks behaves differently than for classical Gaussian Bayesian networks. We establish the parallel between the two theories via tropicalization, and establish the surprising result that the Markov equivalence classes for max-linear Bayesian networks coincide with the ones obtained by regular CI. Our paper opens up many open problems at the intersection of extreme value statistics, causal inference and tropical geometry.

With the rapid increase of digital information we are dealing with in our daily work, we face significant document retrieval and discovery challenges. We present a novel document retrieval and discovery framework that addresses some of the limitations of existing solutions. An innovative aspect of our solution is the combination of implicit and explicit links between documents in the retrieval as well as in the visualisation process, in order to improve document retrieval and discovery. Our framework exploits implicit relationships between documents--defined by the similarity of their content as well as their metadata--and explicit links (hyperlinks) defined between documents based on a third-party link service. Further, the software framework can be extended with arbitrary third-party visualisations. Last but not least, our search query interface offers advanced features not available in most existing document retrieval systems.

We are a year into the development of a software tool for modeling and simulation (M&S) of 1D and 2D kinematics consistent with Newton's laws of motion. Our goal has been to introduce modeling and computational thinking into learning high-school physics. There are two main contributions from an M&S perspective: (1) the use of conceptual modeling, and (2) the application of Finite State Machines (FSMs) to model physical behavior. Both of these techniques have been used by the M&S community to model high-level "soft systems" and discrete events. However, they have not been used to teach physics and represent ways in which M&S can improve physics education. We introduce the NSF-sponsored STEPP project along with its hypothesis and goals. We also describe the development of the three STEPP modules, the server architecture, the assessment plan, and the expected outcomes.