arXiv subjects

Pan Li

Publications and source records attributed to Pan Li.

At least 199 records · Page 11Linked to original sources

arXiv2021

Handling many conversions per click in modeling delayed feedback

Predicting the expected value or number of post-click conversions (purchases or other events) is a key task in performance-based digital advertising. In training a conversion optimizer model, one of the most crucial aspects is handling delayed feedback with respect to conversions, which can happen multiple times with varying delay. This task is difficult, as the delay distribution is different for each advertiser, is long-tailed, often does not follow any particular class of parametric distributions, and can change over time. We tackle these challenges using an unbiased estimation model based on three core ideas. The first idea is to split the label as a sum of labels with different delay buckets, each of which trains only on mature label, the second is to use thermometer encoding to increase accuracy and reduce inference cost, and the third is to use auxiliary information to increase the stability of the model and to handle drift in the distribution.

cs.LG↗

arXiv2020

Revisiting graph neural networks and distance encoding from a practical view

Graph neural networks (GNNs) are widely used in the applications based on graph structured data, such as node classification and link prediction. However, GNNs are often used as a black-box tool and rarely get in-depth investigated regarding whether they fit certain applications that may have various properties. A recently proposed technique distance encoding (DE) (Li et al. 2020) magically makes GNNs work well in many applications, including node classification and link prediction. The theory provided in (Li et al. 2020) supports DE by proving that DE improves the representation power of GNNs. However, it is not obvious how the theory assists the applications accordingly. Here, we revisit GNNs and DE from a more practical point of view. We want to explain how DE makes GNNs fit for node classification and link prediction. Specifically, for link prediction, DE can be viewed as a way to establish correlations between a pair of node representations. For node classification, the problem becomes more complicated as different classification tasks may hold node labels that indicate different physical meanings. We focus on the most widely-considered node classification scenarios and categorize the node labels into two types, community type and structure type, and then analyze different mechanisms that GNNs adopt to predict these two types of labels. We also run extensive experiments to compare eight different configurations of GNNs paired with DE to predict node labels over eight real-world graphs. The results demonstrate the uniform effectiveness of DE to predict structure-type labels. Lastly, we reach three pieces of conclusions on how to use GNNs and DE properly in tasks of node classification.

cs.SI↗

arXiv2020

Distance Encoding: Design Provably More Powerful Neural Networks for Graph Representation Learning

Learning representations of sets of nodes in a graph is crucial for applications ranging from node-role discovery to link prediction and molecule classification. Graph Neural Networks (GNNs) have achieved great success in graph representation learning. However, expressive power of GNNs is limited by the 1-Weisfeiler-Lehman (WL) test and thus GNNs generate identical representations for graph substructures that may in fact be very different. More powerful GNNs, proposed recently by mimicking higher-order-WL tests, only focus on representing entire graphs and they are computationally inefficient as they cannot utilize sparsity of the underlying graph. Here we propose and mathematically analyze a general class of structure-related features, termed Distance Encoding (DE). DE assists GNNs in representing any set of nodes, while providing strictly more expressive power than the 1-WL test. DE captures the distance between the node set whose representation is to be learned and each node in the graph. To capture the distance DE can apply various graph-distance measures such as shortest path distance or generalized PageRank scores. We propose two ways for GNNs to use DEs (1) as extra node features, and (2) as controllers of message aggregation in GNNs. Both approaches can utilize the sparse structure of the underlying graph, which leads to computational efficiency and scalability. We also prove that DE can distinguish node sets embedded in almost all regular graphs where traditional GNNs always fail. We evaluate DE on three tasks over six real networks: structural role prediction, link prediction, and triangle prediction. Results show that our models outperform GNNs without DE by up-to 15\% in accuracy and AUROC. Furthermore, our models also significantly outperform other state-of-the-art methods especially designed for the above tasks.

cs.LG↗

arXiv2020

Quadratic Decomposable Submodular Function Minimization: Theory and Practice (Computation and Analysis of PageRank over Hypergraphs)

We introduce a new convex optimization problem, termed quadratic decomposable submodular function minimization (QDSFM), which allows to model a number of learning tasks on graphs and hypergraphs. The problem exhibits close ties to decomposable submodular function minimization (DSFM), yet is much more challenging to solve. We approach the problem via a new dual strategy and formulate an objective that can be optimized through a number of double-loop algorithms. The outer-loop uses either random coordinate descent (RCD) or alternative projection (AP) methods, for both of which we prove linear convergence rates. The inner-loop computes projections onto cones generated by base polytopes of the submodular functions, via the modified min-norm-point or Frank-Wolfe algorithm. We also describe two new applications of QDSFM: hypergraph-adapted PageRank and semi-supervised learning. The proposed hypergraph-based PageRank algorithm can be used for local hypergraph partitioning, and comes with provable performance guarantees. For hypergraph-adapted semi-supervised learning, we provide numerical experiments demonstrating the efficiency of our QDSFM solvers and their significant improvements on prediction accuracy when compared to state-of-the-art methods.

cs.LG↗

arXiv2020

Graph Information Bottleneck

Representation learning of graph-structured data is challenging because both graph structure and node features carry important information. Graph Neural Networks (GNNs) provide an expressive way to fuse information from network structure and node features. However, GNNs are prone to adversarial attacks. Here we introduce Graph Information Bottleneck (GIB), an information-theoretic principle that optimally balances expressiveness and robustness of the learned representation of graph-structured data. Inheriting from the general Information Bottleneck (IB), GIB aims to learn the minimal sufficient representation for a given task by maximizing the mutual information between the representation and the target, and simultaneously constraining the mutual information between the representation and the input data. Different from the general IB, GIB regularizes the structural as well as the feature information. We design two sampling algorithms for structural regularization and instantiate the GIB principle with two new models: GIB-Cat and GIB-Bern, and demonstrate the benefits by evaluating the resilience to adversarial attacks. We show that our proposed models are more robust than state-of-the-art graph defense models. GIB-based models empirically achieve up to 31% improvement with adversarial perturbation of the graph structure as well as node features.

cs.LG↗

arXiv2020

Latent Unexpected Recommendations

Unexpected recommender system constitutes an important tool to tackle the problem of filter bubbles and user boredom, which aims at providing unexpected and satisfying recommendations to target users at the same time. Previous unexpected recommendation methods only focus on the straightforward relations between current recommendations and user expectations by modeling unexpectedness in the feature space, thus resulting in the loss of accuracy measures in order to improve unexpectedness performance. Contrast to these prior models, we propose to model unexpectedness in the latent space of user and item embeddings, which allows to capture hidden and complex relations between new recommendations and historic purchases. In addition, we develop a novel Latent Closure (LC) method to construct hybrid utility function and provide unexpected recommendations based on the proposed model. Extensive experiments on three real-world datasets illustrate superiority of our proposed approach over the state-of-the-art unexpected recommendation models, which leads to significant increase in unexpectedness measure without sacrificing any accuracy metric under all experimental settings in this paper.

cs.IR↗

arXiv2020

Reflected continuously tunable acoustic metasurface with rotatable space coiling-up structure

In this paper, we propose a continuously tunable acoustic metasurface composed of identical anisotropic resonant units, each of which contains a rigid pedestal and a rotatable inclusion with space coiling-up structure. The metasurface can manipulate the reflected phase by adjusting the rotational angle of inclusion. The theoretical analysis shows that the polarization-dependent phase change can be induced by the even-order standing wave modes inside inclusion. By utilizing the rotatable inclusion, we design a tunable acoustic carpet cloaking device, which works with a wide range for incident angle. When incident waves come from different directions, the cloaking effect can be obtained by arrange the rotational angle of each inclusion.

physics.app-ph↗

arXiv2020

Towards Controllable and Personalized Review Generation

In this paper, we propose a novel model RevGAN that automatically generates controllable and personalized user reviews based on the arbitrarily given sentimental and stylistic information. RevGAN utilizes the combination of three novel components, including self-attentive recursive autoencoders, conditional discriminators, and personalized decoders. We test its performance on the several real-world datasets, where our model significantly outperforms state-of-the-art generation models in terms of sentence quality, coherence, personalization and human evaluations. We also empirically show that the generated reviews could not be easily distinguished from the organically produced reviews and that they follow the same statistical linguistics laws.

cs.CL↗

arXiv2019

Optimizing Generalized PageRank Methods for Seed-Expansion Community Detection

Landing probabilities (LP) of random walks (RW) over graphs encode rich information regarding graph topology. Generalized PageRanks (GPR), which represent weighted sums of LPs of RWs, utilize the discriminative power of LP features to enable many graph-based learning studies. Previous work in the area has mostly focused on evaluating suitable weights for GPRs, and only a few studies so far have attempted to derive the optimal weights of GRPs for a given application. We take a fundamental step forward in this direction by using random graph models to better our understanding of the behavior of GPRs. In this context, we provide a rigorous non-asymptotic analysis for the convergence of LPs and GPRs to their mean-field values on edge-independent random graphs. Although our theoretical results apply to many problem settings, we focus on the task of seed-expansion community detection over stochastic block models. There, we find that the predictive power of LPs decreases significantly slower than previously reported based on asymptotic findings. Given this result, we propose a new GPR, termed Inverse PR (IPR), with LP weights that increase for the initial few steps of the walks. Extensive experiments on both synthetic and real, large-scale networks illustrate the superiority of IPR compared to other GPRs for seeded community detection.

cs.SI↗

arXiv2019

Continuously Tunable Acoustic Metasurface with Rotatable Anisotropic Three-component Resonators

We propose a tunable acoustic metasurface consisting of identical units. And units are rotatable anisotropic three-component resonators, which can induce the non-degenerate dipolar resonance, causing an evident phase change in low frequencies. Compared with the monopole resonance widely used in Helmholtz resonators, the polarization direction of the dipole resonance is a new degree of freedom for phase manipulation. The proposed metasurface is constructed by identical units that made with real (not rigid) materials. And the phase profile can continuously change by rotating the anisotropic resonators. We present a wide-angle and broad-band acoustic focusing by the metasurface under a water background.

physics.app-ph↗

arXiv2019

DDTCDR: Deep Dual Transfer Cross Domain Recommendation

Cross domain recommender systems have been increasingly valuable for helping consumers identify the most satisfying items from different categories. However, previously proposed cross-domain models did not take into account bidirectional latent relations between users and items. In addition, they do not explicitly model information of user and item features, while utilizing only user ratings information for recommendations. To address these concerns, in this paper we propose a novel approach to cross-domain recommendations based on the mechanism of dual learning that transfers information between two related domains in an iterative manner until the learning process stabilizes. We develop a novel latent orthogonal mapping to extract user preferences over multiple domains while preserving relations between users across different latent spaces. Combining with autoencoder approach to extract the latent essence of feature information, we propose Deep Dual Transfer Cross Domain Recommendation (DDTCDR) model to provide recommendations in respective domains. We test the proposed method on a large dataset containing three domains of movies, book and music items and demonstrate that it consistently and significantly outperforms several state-of-the-art baselines and also classical transfer learning approaches.

cs.IR↗

arXiv2019

A Capacity-Price Game for Uncertain Renewables Resources

Renewable resources are starting to constitute a growing portion of the total generation mix of the power system. A key difference between renewables and traditional generators is that many renewable resources are managed by individuals, especially in the distribution system. In this paper, we study the capacity investment and pricing problem, where multiple renewable producers compete in a decentralized market. It is known that most deterministic capacity games tend to result in very inefficient equilibria, even when there are a large number of similar players. In contrast, we show that due to the inherent randomness of renewable resources, the equilibria in our capacity game becomes efficient as the number of players grows and coincides with the centralized decision from the social planner's problem. This result provides a new perspective on how to look at the positive influence of randomness in a game framework as well as its contribution to resource planning, scheduling, and bidding. We validate our results by simulation studies using real world data.

math.OC↗

arXiv2019

Meta-Graph Based HIN Spectral Embedding: Methods, Analyses, and Insights

In this work, we propose to study the utility of different meta-graphs, as well as how to simultaneously leverage multiple meta-graphs for HIN embedding in an unsupervised manner. Motivated by prolific research on homogeneous networks, especially spectral graph theory, we firstly conduct a systematic empirical study on the spectrum and embedding quality of different meta-graphs on multiple HINs, which leads to an efficient method of meta-graph assessment. It also helps us to gain valuable insight into the higher-order organization of HINs and indicates a practical way of selecting useful embedding dimensions. Further, we explore the challenges of combining multiple meta-graphs to capture the multi-dimensional semantics in HIN through reasoning from mathematical geometry and arrive at an embedding compression method of autoencoder with $\ell_{2,1}$-loss, which finds the most informative meta-graphs and embeddings in an end-to-end unsupervised manner. Finally, empirical analysis suggests a unified workflow to close the gap between our meta-graph assessment and combination methods. To the best of our knowledge, this is the first research effort to provide rich theoretical and empirical analyses on the utility of meta-graphs and their combinations, especially regarding HIN embedding. Extensive experimental comparisons with various state-of-the-art neural network based embedding methods on multiple real-world HINs demonstrate the effectiveness and efficiency of our framework in finding useful meta-graphs and generating high-quality HIN embeddings.

cs.LG↗

arXiv2019

Ultra-thin Underwater Acoustic Metasurface with Multiply Resonant Units

This paper describes a new kind of acoustic metasurface with multiply resonant units, which have previously been used to induce multiple resonances and effectively produce negative mass density and bulk/shear moduli. The proposed acoustic metasurface can be constructed using real materials and does not rely on an ideal rigid material. Therefore, it can work well in a water background. The thickness of the acoustic metasurface is about two orders of magnitude smaller than the acoustic wavelength in water. The design of a unit group is proposed to avoid the phase discretization becoming too fine in such a long-wavelength condition. We demonstrate that the proposed acoustic metasurface achieves good performance in anomalous reflection, focusing, and carpet cloaking.

physics.app-ph↗

arXiv2019

Subspace Determination through Local Intrinsic Dimensional Decomposition: Theory and Experimentation

Axis-aligned subspace clustering generally entails searching through enormous numbers of subspaces (feature combinations) and evaluation of cluster quality within each subspace. In this paper, we tackle the problem of identifying subsets of features with the most significant contribution to the formation of the local neighborhood surrounding a given data point. For each point, the recently-proposed Local Intrinsic Dimension (LID) model is used in identifying the axis directions along which features have the greatest local discriminability, or equivalently, the fewest number of components of LID that capture the local complexity of the data. In this paper, we develop an estimator of LID along axis projections, and provide preliminary evidence that this LID decomposition can indicate axis-aligned data subspaces that support the formation of clusters.

cs.LG↗

arXiv2019

Latent Multi-Criteria Ratings for Recommendations

Multi-criteria recommender systems have been increasingly valuable for helping consumers identify the most relevant items based on different dimensions of user experiences. However, previously proposed multi-criteria models did not take into account latent embeddings generated from user reviews, which capture latent semantic relations between users and items. To address these concerns, we utilize variational autoencoders to map user reviews into latent embeddings, which are subsequently compressed into low-dimensional discrete vectors. The resulting compressed vectors constitute latent multi-criteria ratings that we use for the recommendation purposes via standard multi-criteria recommendation methods. We show that the proposed latent multi-criteria rating approach outperforms several baselines significantly and consistently across different datasets and performance evaluation measures.

cs.LG↗

arXiv2019

Latent Unexpected and Useful Recommendation

Providing unexpected recommendations is an important task for recommender systems. To do this, we need to start from the expectations of users and deviate from these expectations when recommending items. Previously proposed approaches model user expectations in the feature space, making them limited to the items that the user has visited or expected by the deduction of associated rules, without including the items that the user could also expect from the latent, complex and heterogeneous interactions between users, items and entities. In this paper, we define unexpectedness in the latent space rather than in the feature space and develop a novel Latent Convex Hull (LCH) method to provide unexpected recommendations. Extensive experiments on two real-world datasets demonstrate the effectiveness of the proposed model that significantly outperforms alternative state-of-the-art unexpected recommendation methods in terms of unexpectedness measures while achieving the same level of accuracy.

cs.IR↗

arXiv2019

Scaling-up atomically thin coplanar semiconductor-metal circuitry via phase engineered chemical assembly

Two-dimensional (2D) layered semiconductors, with their ultimate atomic thickness, have shown promise to scale down transistors for modern integrated circuitry. However, the electrical contacts that connect these materials with external bulky metals are usually unsatisfactory, which limits the transistor performance. Recently, contacting 2D semiconductors using coplanar 2D conductors has shown promise in reducing the problematic high resistance contacts. However, many of these methods are not ideal for scaled production. Here, we report on the large-scale, spatially controlled chemical assembly of the integrated 2H-MoTe2 field-effect transistors (FETs) with coplanar metallic 1T' MoTe2 contacts via phase engineered approaches. We demonstrate that the heterophase FETs exhibit ohmic contact behavior with low contact resistance, resulting from the coplanar seamless contact between 2H and 1T' MoTe2 confirmed by transmission electron microscopy characterizations. The average mobility of the heterophase FETs was measured to be as high as 23 cm2 V-1 s-1 (comparable with those of exfoliated single crystals), due to the large 2H MoTe2 single-crystalline domain (486μm). By developing a patterned growth method, we realize the 1T' MoTe2 gated heterophase FET array whose components of channel, gate, and contacts are all 2D materials. Finally, we transfer the heterophase device array onto a flexible substrate and demonstrate the near-infrared photoresponse with high photoresponsivity (~1.02 A/W). Our study provides a basis for the large-scale application of phase-engineered coplanar MoTe2 semiconductors-meter structure in advanced electronics and optoelectronics.

cond-mat.mtrl-sci↗