A Unified Framework for Cross-Domain and Cross-System Recommendations

Cross-Domain Recommendation (CDR) and Cross-System Recommendation (CSR) have been proposed to improve the recommendation accuracy in a target dataset (domain/system) with the help of a source one with relatively richer information. However, most existing CDR and CSR approaches are single-target, namely, there is a single target dataset, which can only help the target dataset and thus cannot benefit the source dataset. In this paper, we focus on three new scenarios, i.e., Dual-Target CDR (DTCDR), Multi-Target CDR (MTCDR), and CDR+CSR, and aim to improve the recommendation accuracy in all datasets simultaneously for all scenarios. To do this, we propose a unified framework, called GA (based on Graph embedding and Attention techniques), for all three scenarios. In GA, we first construct separate heterogeneous graphs to generate more representative user and item embeddings. Then, we propose an element-wise attention mechanism to effectively combine the embeddings of common entities (users/items) learned from different datasets. Moreover, to avoid negative transfer, we further propose a Personalized training strategy to minimize the embedding difference of common entities between a richer dataset and a sparser dataset, deriving three new models, i.e., GA-DTCDR-P, GA-MTCDR-P, and GA-CDR+CSR-P, for the three scenarios respectively. Extensive experiments conducted on four real-world datasets demonstrate that our proposed GA models significantly outperform the state-of-the-art approaches.

Subnet Replacement: Deployment-stage backdoor attack against deep neural networks in gray-box setting

We study the realistic potential of conducting backdoor attack against deep neural networks (DNNs) during deployment stage. Specifically, our goal is to design a deployment-stage backdoor attack algorithm that is both threatening and realistically implementable. To this end, we propose Subnet Replacement Attack (SRA), which is capable of embedding backdoor into DNNs by directly modifying a limited number of model parameters. Considering the realistic practicability, we abandon the strong white-box assumption widely adopted in existing studies, instead, our algorithm works in a gray-box setting, where architecture information of the victim model is available but the adversaries do not have any knowledge of parameter values. The key philosophy underlying our approach is -- given any neural network instance (regardless of its specific parameter values) of a certain architecture, we can always embed a backdoor into that model instance, by replacing a very narrow subnet of a benign model (without backdoor) with a malicious backdoor subnet, which is designed to be sensitive (fire large activation value) to a particular backdoor trigger pattern.

A new method for vehicle system safety design based on data mining with uncertainty modeling

In this research, a new data mining-based design approach has been developed for designing complex mechanical systems such as a crashworthy passenger car with uncertainty modeling. The method allows exploring the big crash simulation dataset to design the vehicle at multi-levels in a top-down manner (main energy absorbing system, components, and geometric features) and derive design rules based on the whole vehicle body safety requirements to make decisions towards the component and sub-component level design. Full vehicle and component simulation datasets are mined to build decision trees, where the interrelationship among parameters can be revealed and the design rules are derived to produce designs with good performance. This method has been extended by accounting for the uncertainty in the design variables. A new decision tree algorithm for uncertain data (DTUD) is developed to produce the desired designs and evaluate the design performance variations due to the uncertainty in design variables. The framework of this method is implemented by combining the design of experiments (DOE) and crash finite element analysis (FEA) and then demonstrated by designing a passenger car subject to front impact. The results show that the new methodology could achieve the design objectives efficiently and effectively. By applying the new method, the reliability of the final designs is also increased greatly. This approach has the potential to be applied as a general design methodology for a wide range of complex structures and mechanical systems.

Topological magnon insulators in two-dimensional van der Waals ferromagnets CrSiTe$_3$ and CrGeTe$_3$: towards intrinsic gap-tunability

The bosonic analogues of topological insulators have been proposed in numerous theoretical works, but their experimental realization is still very rare, especially for spin systems. Recently, two-dimensional (2D) honeycomb van der Waals (vdW) ferromagnets have emerged as a new platform for topological spin excitations. Here, via a comprehensive inelastic neutron scattering study and theoretical analysis of the spin-wave excitations, we report the realization of topological magnon insulators in CrXTe$_3$ (X=Si, Ge) compounds. The nontrivial nature and intrinsic tunability of the gap opening at the magnon band-crossing Dirac points are confirmed, while the emergence of the corresponding in-gap topological edge states is demonstrated theoretically. The realization of topological magnon insulators with intrinsic gap-tunability in this class of remarkable 2D materials will undoubtedly lead to new and fascinating technological applications in the domain of magnonics and topological spintronics.

Dynamic Planning and Learning under Recovering Rewards

Motivated by emerging applications such as live-streaming e-commerce, promotions and recommendations, we introduce a general class of multi-armed bandit problems that have the following two features: (i) the decision maker can pull and collect rewards from at most $K$ out of $N$ different arms in each time period; (ii) the expected reward of an arm immediately drops after it is pulled, and then non parametrically recovers as the idle time increases. With the objective of maximizing expected cumulative rewards over $T$ time periods, we propose, construct and prove performance guarantees for a class of Purely Periodic Policies. For the offline problem when all model parameters are known, our proposed policy obtains an approximation ratio that is at the order of $1-mathcal O(1/sqrt{K})$, which is asymptotically optimal when $K$ grows to infinity. For the online problem when the model parameters are unknown and need to be learned, we design an Upper Confidence Bound (UCB) based policy that approximately has $widetilde{mathcal O}(Nsqrt{T})$ regret against the offline benchmark. Our framework and policy design may have the potential to be adapted into other offline planning and online learning applications with non-stationary and recovering rewards.

Spatiotemporal differentiators generating optical vortices with pure transverse orbital angular momentum and detecting sharp change of pulse envelope

As a new degree of freedom for optical manipulation, recently spatiotemporal optical vortices (STOVs) carrying transverse orbital angular momentums have been experimentally demonstrated with bulky optical systems. Here we propose a spatiotemporal differentiator to generate STOVs with pure transverse orbital angular momentum. In order to create phase singularity in the spatiotemporal domain, we design a spatiotemporal differentiator by breaking spatial mirror symmetry. In contrast to the complex bulky systems, the device we propose here is a simple one-dimensional periodic nanostructure and thus it is much more compact. We show that for a normal incident pulse, the differentiator generates a transmitted STOV pulse with transverse orbital angular momentum. Furthermore, we demonstrate that the interference of the generated STOVs can be used to detect the sharp changes of pulse envelopes, in both spatial and temporal dimensions.

Ergodicity and equidistribution in Hilbert geometry

We show that dynamical and counting results characteristic of negatively-curved Riemannian geometry, or more generally CAT($-1$) or rank-one CAT(0) spaces, also hold for rank-one properly convex projective structures, equipped with their Hilbert metrics, admitting finite Sullivan measures built from appropriate conformal densities. In particular, this includes geometrically finite convex projective structures. More specifically, with respect to the Sullivan measure, the Hilbert geodesic flow is strongly mixing, and orbits and primitive closed geodesics equidistribute, allowing us to asymptotically enumerate these objects.

FedScale: Benchmarking Model and System Performance of Federated Learning

We present FedScale, a diverse set of challenging and realistic benchmark datasets to facilitate scalable, comprehensive, and reproducible federated learning (FL) research. FedScale datasets are large-scale, encompassing a diverse range of important FL tasks, such as image classification, object detection, language modeling, speech recognition, and reinforcement learning. For each dataset, we provide a unified evaluation protocol using realistic data splits and evaluation metrics. To meet the pressing need for reproducing realistic FL at scale, we have also built an efficient evaluation platform to simplify and standardize the process of FL experimental setup and model evaluation. Our evaluation platform provides flexible APIs to implement new FL algorithms and includes new execution backends with minimal developer efforts. Finally, we perform indepth benchmark experiments on these datasets. Our experiments suggest fruitful opportunities in heterogeneity-aware co-optimizations of the system and statistical efficiency under realistic FL characteristics. FedScale is open-source with permissive licenses and actively maintained,1 and we welcome feedback and contributions from the community.

Synergies Between Affordance and Geometry: 6-DoF Grasp Detection via Implicit Representations

Grasp detection in clutter requires the robot to reason about the 3D scene from incomplete and noisy perception. In this work, we draw insight that 3D reconstruction and grasp learning are two intimately connected tasks, both of which require a fine-grained understanding of local geometry details. We thus propose to utilize the synergies between grasp affordance and 3D reconstruction through multi-task learning of a shared representation. Our model takes advantage of deep implicit functions, a continuous and memory-efficient representation, to enable differentiable training of both tasks. We train the model on self-supervised grasp trials data in simulation. Evaluation is conducted on a clutter removal task, where the robot clears cluttered objects by grasping them one at a time. The experimental results in simulation and on the real robot have demonstrated that the use of implicit neural representations and joint learning of grasp affordance and 3D reconstruction have led to state-of-the-art grasping results. Our method outperforms baselines by over 10% in terms of grasp success rate. Additional results and videos can be found at https://sites.google.com/view/rpl-giga2021

Detailed analysis of the poorly studied northern open cluster NGC 1348 using multi-color photometry and GAIA EDR3 astrometry

The membership determination for open clusters in noisy environments of the Milky Way is still an open problem. In this paper, our main aim is provide the membership probability of stars using proper motions and parallax values of stars using Gaia EDR3 astrometry. Apart from the Gaia astrometry, we have also used other photometric data sets like UKIDSS, WISE, APASS and Pan-STARRS1 in order to understand cluster properties from optical to mid-infrared regions. We selected 438 likely members with membership probability higher than $50%$ and G$le$20 mag. We obtained the mean value of proper motion as $mu_{x}=1.27pm0.001$ and $mu_{y}=-0.73pm0.002$ mas yr$^{-1}$. The clusters radius is determined as 7.5 arcmin (5.67 pc) using radial density profile. Our analysis suggests that NGC 1348 is located at a distance of $2.6pm0.05$ kpc. The mass function slope is found to be $1.30pm0.18$ in the mass range 1.0$-$4.1 $M_odot$, which is in fair agreement with Salpeters value within the 1$sigma$ uncertainty. The present study validates that NGC 1348 is a dynamically relaxed cluster. We computed the apex coordinates $(A, D)$ for NGC 1348 as $(A_circ, D_circ)$ = $(-23^{textrm{o}}.815pm 0^{textrm{o}}.135$, $-22^{textrm{o}}.228pm 0^{textrm{o}}.105)$. In addition, calculations of the velocity ellipsoid parameters (VEPs), matrix elements $mu_{ij}$, direction cosines ($l_j$, $m_j$, $n_j$) and the Galactic longitude of the vertex have been also conducted in this analysis.