arXiv subjects

Lei Chen

Publications and source records attributed to Lei Chen.

At least 523 records · Page 29Linked to original sources

arXiv2021

Adaptive Appearance Rendering

We propose an approach to generate images of people given a desired appearance and pose. Disentangled representations of pose and appearance are necessary to handle the compound variability in the resulting generated images. Hence, we develop an approach based on intermediate representations of poses and appearance: our pose-guided appearance rendering network firstly encodes the targets' poses using an encoder-decoder neural network. Then the targets' appearances are encoded by learning adaptive appearance filters using a fully convolutional network. Finally, these filters are placed in the encoder-decoder neural networks to complete the rendering. We demonstrate that our model can generate images and videos that are superior to state-of-the-art methods, and can handle pose guided appearance rendering in both image and video generation.

cs.CV↗

arXiv2021

Piggyback GAN: Efficient Lifelong Learning for Image Conditioned Generation

Humans accumulate knowledge in a lifelong fashion. Modern deep neural networks, on the other hand, are susceptible to catastrophic forgetting: when adapted to perform new tasks, they often fail to preserve their performance on previously learned tasks. Given a sequence of tasks, a naive approach addressing catastrophic forgetting is to train a separate standalone model for each task, which scales the total number of parameters drastically without efficiently utilizing previous models. In contrast, we propose a parameter efficient framework, Piggyback GAN, which learns the current task by building a set of convolutional and deconvolutional filters that are factorized into filters of the models trained on previous tasks. For the current task, our model achieves high generation quality on par with a standalone model at a lower number of parameters. For previous tasks, our model can also preserve generation quality since the filters for previous tasks are not altered. We validate Piggyback GAN on various image-conditioned generation tasks across different domains, and provide qualitative and quantitative results to show that the proposed approach can address catastrophic forgetting effectively and efficiently.

cs.CV↗

arXiv2021

Learning Fair Representations for Recommendation: A Graph-based Perspective

As a key application of artificial intelligence, recommender systems are among the most pervasive computer aided systems to help users find potential items of interests. Recently, researchers paid considerable attention to fairness issues for artificial intelligence applications. Most of these approaches assumed independence of instances, and designed sophisticated models to eliminate the sensitive information to facilitate fairness. However, recommender systems differ greatly from these approaches as users and items naturally form a user-item bipartite graph, and are collaboratively correlated in the graph structure. In this paper, we propose a novel graph based technique for ensuring fairness of any recommendation models. Here, the fairness requirements refer to not exposing sensitive feature set in the user modeling process. Specifically, given the original embeddings from any recommendation models, we learn a composition of filters that transform each user's and each item's original embeddings into a filtered embedding space based on the sensitive feature set. For each user, this transformation is achieved under the adversarial learning of a user-centric graph, in order to obfuscate each sensitive feature between both the filtered user embedding and the sub graph structures of this user. Finally, extensive experimental results clearly show the effectiveness of our proposed model for fair recommendation. We publish the source code at https://github.com/newlei/FairGo.

cs.IR↗

arXiv2021

Generalization of Wigner Time Delay to Sub-Unitary Scattering Systems

We introduce a complex generalization of Wigner time delay $τ$ for sub-unitary scattering systems. Theoretical expressions for complex time delay as a function of excitation energy, uniform and non-uniform loss, and coupling, are given. We find very good agreement between theory and experimental data taken on microwave graphs containing an electronically variable lumped-loss element. We find that time delay and the determinant of the scattering matrix share a common feature in that the resonant behavior in $\text{Re}[τ]$ and $\text{Im}[τ]$ serves as a reliable indicator of the condition for Coherent Perfect Absorption (CPA). This work opens a new window on time delay in lossy systems and provides a means to identify the poles and zeros of the scattering matrix from experimental data. The results also enable a new approach to achieving CPA at an arbitrary energy/frequency in complex scattering systems.

cond-mat.dis-nn↗

arXiv2021

Efficient Relation-aware Scoring Function Search for Knowledge Graph Embedding

The scoring function, which measures the plausibility of triplets in knowledge graphs (KGs), is the key to ensure the excellent performance of KG embedding, and its design is also an important problem in the literature. Automated machine learning (AutoML) techniques have recently been introduced into KG to design task-aware scoring functions, which achieve state-of-the-art performance in KG embedding. However, the effectiveness of searched scoring functions is still not as good as desired. In this paper, observing that existing scoring functions can exhibit distinct performance on different semantic patterns, we are motivated to explore such semantics by searching relation-aware scoring functions. But the relation-aware search requires a much larger search space than the previous one. Hence, we propose to encode the space as a supernet and propose an efficient alternative minimization algorithm to search through the supernet in a one-shot manner. Finally, experimental results on benchmark datasets demonstrate that the proposed method can efficiently search relation-aware scoring functions, and achieve better embedding performance than state-of-the-art methods.

cs.LG↗

arXiv2021

Searching to Sparsify Tensor Decomposition for N-ary Relational Data

Tensor, an extension of the vector and matrix to the multi-dimensional case, is a natural way to describe the N-ary relational data. Recently, tensor decomposition methods have been introduced into N-ary relational data and become state-of-the-art on embedding learning. However, the performance of existing tensor decomposition methods is not as good as desired. First, they suffer from the data-sparsity issue since they can only learn from the N-ary relational data with a specific arity, i.e., parts of common N-ary relational data. Besides, they are neither effective nor efficient enough to be trained due to the over-parameterization problem. In this paper, we propose a novel method, i.e., S2S, for effectively and efficiently learning from the N-ary relational data. Specifically, we propose a new tensor decomposition framework, which allows embedding sharing to learn from facts with mixed arity. Since the core tensors may still suffer from the over-parameterization, we propose to reduce parameters by sparsifying the core tensors while retaining their expressive power using neural architecture search (NAS) techniques, which can search for data-dependent architectures. As a result, the proposed S2S not only guarantees to be expressive but also efficiently learns from mixed arity. Finally, empirical results have demonstrated that S2S is efficient to train and achieves state-of-the-art performance.

cs.LG↗

arXiv2021

Improving the Precision of Optical Metrology by Detecting Fewer Photons

In optical metrological protocols to measure physical quantities, it is, in principle, always beneficial to increase photon number to improve measurement precision. However, practical constraints prevent arbitrary increase of n due to the imperfections of a practical detector, especially when the detector response is dominated by saturation effect. In this work, we show that a modified weak measurement protocol, namely, biased weak measurement significantly improves the precision of optical metrology in the presence of saturation effect. This method detects an ultra-small fraction of photons while maintains considerable amount of metrological information. The biased pre-coupling leads to an additional reduction of photons in the post-selection and generates an extinction point in the spectrum distribution, which is extremely sensitive to the estimated parameter and difficult to be saturated. Therefore, the Fisher information can be persistently enhanced by increasing the photon number. In our magnetic-sensing experiment, biased weak measurement achieves precision approximately one order of magnitude better than those of previously used methods. The proposed method can be applied in various optical measurement schemes to circumvent detector saturation effect with low-cost apparatuses.

quant-ph↗

arXiv2021

ZYELL-NCTU NetTraffic-1.0: A Large-Scale Dataset for Real-World Network Anomaly Detection

Network security has been an active research topic for long. One critical issue is improving the anomaly detection capability of intrusion detection systems (IDSs), such as firewalls. However, existing network anomaly datasets are out of date (i.e., being collected many years ago) or IP-anonymized, making the data characteristics differ from today's network. Therefore, this work introduces a new, large-scale, and real-world dataset, ZYELL-NCTU NetTraffic-1.0, which is collected from the raw output of firewalls in a real network, with the objective to advance the development of network security researches.

cs.CR↗

arXiv2021

Revealing the intrinsic superconducting gap anisotropy in surface-neutralized BaFe$_2$(As$_{0.7}$P$_{0.3}$)$_2$

Alkaline-earth iron arsenide (122) is one of the most studied families of iron-based superconductors, especially for angle-resolved photoemission spectroscopy. While extensive photoemission results have been obtained, the surface complexity of 122 caused by its charge-non-neutral surface is rarely considered. Here, we show that the surface of 122 can be neutralized by potassium deposition. In potassium-coated BaFe$_2$(As$_{0.7}$P$_{0.3}$)$_2$, the surface-induced spectral broadening is strongly suppressed, and hence the coherent spectra that reflect the intrinsic bulk electronic state recover. This enables the measuring of superconducting gap with unpreceded precision. The result shows the existence of two pairing channels. While the gap anisotropy on the outer hole/electron pockets can be well fitted using an s$_\pm$ gap function, the gap anisotropy on the inner hole/electron shows a clear deviation. Our results provide quantitative constraints for refining theoretical models and also demonstrate an experimental method for revealing the intrinsic electronic properties of 122 in future studies.

cond-mat.supr-con↗

arXiv2021

Field-free spin-orbit torque-induced switching of perpendicular magnetization in a ferrimagnetic layer with vertical composition gradient

Current-induced spin-orbit torques (SOTs) are of interest for fast and energy-efficient manipulation of magnetic order in spintronic devices. To be deterministic, however, switching of perpendicularly magnetized materials by SOT requires a mechanism for in-plane symmetry breaking. Existing methods to do so involve the application of an in-plane bias magnetic field, or incorporation of in-plane structural asymmetry in the device, both of which can be difficult to implement in practical applications. Here, we reported bias-field-free SOT switching in a single perpendicular CoTb layer with an engineered vertical composition gradient. The vertical structural inversion asymmetry induces strong intrinsic SOTs and a gradient-driven Dzyaloshinskii-Moriya interaction (g-DMI), which breaks the in-plane symmetry during the switching process. Micromagnetic simulations are in agreement with experimental results, and elucidate the role of g-DMI in the deterministic switching. This bias-field-free switching scheme for perpendicular ferrimagnets with g-DMI provides a strategy for efficient and compact SOT device design.

cond-mat.mes-hall↗

arXiv2021

Observation of robust edge superconductivity in Fe(Se,Te) under strong magnetic perturbation

The iron-chalcogenide high temperature superconductor Fe(Se,Te) (FST) has been reported to exhibit complex magnetic ordering and nontrivial band topology which may lead to novel superconducting phenomena. However, the recent studies have so far been largely concentrated on its band and spin structures while its mesoscopic electronic and magnetic response, crucial for future device applications, has not been explored experimentally. Here, we used scanning superconducting quantum interference device microscopy for its sensitivity to both local diamagnetic susceptibility and current distribution in order to image the superfluid density and supercurrent in FST. We found that in FST with 10% interstitial Fe, whose magnetic structure was heavily disrupted, bulk superconductivity was significantly suppressed whereas edge still preserved strong superconducting diamagnetism. The edge dominantly carried supercurrent despite of a very long magnetic penetration depth. The temperature dependence of the superfluid density and supercurrent distribution were distinctively different between the edge and the bulk. Our Heisenberg modeling showed that magnetic dopants stabilize anti-ferromagnetic spin correlation along the edge, which may contribute towards its robust superconductivity. Our observations hold implication for FST as potential platforms for topological quantum computation and superconducting spintronics.

cond-mat.supr-con↗

arXiv2021

Far-Field Super-Resolution Imaging By Nonlinear Excited Evanescent Waves

Abbe's resolution limit, one of the best-known physical limitations, poses a great challenge for any wave systems in imaging, wave transport, and dynamics. Originally formulated in linear optics, this Abbe's limit can be broken using nonlinear optical interactions. Here we extend the Abbe theory into a nonlinear regime and experimentally demonstrate a far-field, label-free, and scan-free super-resolution imaging technique based on nonlinear four-wave mixing to retrieve near-field scattered evanescent waves, achieving sub-wavelength resolution of $λ/15.6$. This method paves the way for application in biomedical imaging, semiconductor metrology, and photolithography.

physics.optics↗

arXiv2021

UniNet: Scalable Network Representation Learning with Metropolis-Hastings Sampling

Network representation learning (NRL) technique has been successfully adopted in various data mining and machine learning applications. Random walk based NRL is one popular paradigm, which uses a set of random walks to capture the network structural information, and then employs word2vec models to learn the low-dimensional representations. However, until now there is lack of a framework, which unifies existing random walk based NRL models and supports to efficiently learn from large networks. The main obstacle comes from the diverse random walk models and the inefficient sampling method for the random walk generation. In this paper, we first introduce a new and efficient edge sampler based on Metropolis-Hastings sampling technique, and theoretically show the convergence property of the edge sampler to arbitrary discrete probability distributions. Then we propose a random walk model abstraction, in which users can easily define different transition probability by specifying dynamic edge weights and random walk states. The abstraction is efficiently supported by our edge sampler, since our sampler can draw samples from unnormalized probability distribution in constant time complexity. Finally, with the new edge sampler and random walk model abstraction, we carefully implement a scalable NRL framework called UniNet. We conduct comprehensive experiments with five random walk based NRL models over eleven real-world datasets, and the results clearly demonstrate the efficiency of UniNet over billion-edge networks.

cs.LG↗

arXiv2020

Quantitative Evaluations on Saliency Methods: An Experimental Study

It has been long debated that eXplainable AI (XAI) is an important topic, but it lacks rigorous definition and fair metrics. In this paper, we briefly summarize the status quo of the metrics, along with an exhaustive experimental study based on them, including faithfulness, localization, false-positives, sensitivity check, and stability. With the experimental results, we conclude that among all the methods we compare, no single explanation method dominates others in all metrics. Nonetheless, Gradient-weighted Class Activation Mapping (Grad-CAM) and Randomly Input Sampling for Explanation (RISE) perform fairly well in most of the metrics. Utilizing a set of filtered metrics, we further present a case study to diagnose the classification bases for models. While providing a comprehensive experimental study of metrics, we also examine measuring factors that are missed in current metrics and hope this valuable work could serve as a guide for future research.

cs.AI↗

arXiv2020

yNet: a multi-input convolutional network for ultra-fast simulation of field evolvement

The capability of multi-input field-to-field regression, i.e. mapping the initial field and applied conditions to the evolved field, is appealing, enabling ultra-fast physics-free simulation of various field evolvements across many disciplines. We hereby propose a y-shaped multi-input deep convolutional network, yNet, which can effectively account for combined effects of multiple mixed inputs on output field. The proposed yNet is applied to the simulation of porosity evolution in selective lasering sintering (SLS). Upon testing, yNet can simulate nearly identical porosity evolution and development to the physics-based model, with a 99.13% morphological similarity for various SLS conditions. We then effortlessly boost the porosity simulation capability to the realistic, full-component level. yNet is generally applicable to simulating various structural/morphological evolutions and other condition-concerned, continuous field evolvements even with spatially and/or temporally non-uniform evolving kinetics. Once trained, the light-weight yNet can be distributed easily and ran with limited computational resource while reducing computation time to a split second. It thus may have a transformative impact by democratizing the capability of ultra-fast and extreme-scale simulation of various field evolvements.

cs.CE↗

arXiv2020

On Graph Neural Networks versus Graph-Augmented MLPs

From the perspective of expressive power, this work compares multi-layer Graph Neural Networks (GNNs) with a simplified alternative that we call Graph-Augmented Multi-Layer Perceptrons (GA-MLPs), which first augments node features with certain multi-hop operators on the graph and then applies an MLP in a node-wise fashion. From the perspective of graph isomorphism testing, we show both theoretically and numerically that GA-MLPs with suitable operators can distinguish almost all non-isomorphic graphs, just like the Weifeiler-Lehman (WL) test. However, by viewing them as node-level functions and examining the equivalence classes they induce on rooted graphs, we prove a separation in expressive power between GA-MLPs and GNNs that grows exponentially in depth. In particular, unlike GNNs, GA-MLPs are unable to count the number of attributed walks. We also demonstrate via community detection experiments that GA-MLPs can be limited by their choice of operator family, as compared to GNNs with higher flexibility in learning.

cs.LG↗

arXiv2020

Scalable Federated Learning over Passive Optical Networks

Two-step aggregation is introduced to facilitate scalable federated learning (SFL) over passive optical networks (PONs). Results reveal that the SFL keeps the required PON upstream bandwidth constant regardless of the number of involved clients, while bringing ~10% learning accuracy improvement.

cs.NI↗

arXiv2020

Can Graph Neural Networks Count Substructures?

The ability to detect and count certain substructures in graphs is important for solving many tasks on graph-structured data, especially in the contexts of computational chemistry and biology as well as social network analysis. Inspired by this, we propose to study the expressive power of graph neural networks (GNNs) via their ability to count attributed graph substructures, extending recent works that examine their power in graph isomorphism testing and function approximation. We distinguish between two types of substructure counting: induced-subgraph-count and subgraph-count, and establish both positive and negative answers for popular GNN architectures. Specifically, we prove that Message Passing Neural Networks (MPNNs), 2-Weisfeiler-Lehman (2-WL) and 2-Invariant Graph Networks (2-IGNs) cannot perform induced-subgraph-count of substructures consisting of 3 or more nodes, while they can perform subgraph-count of star-shaped substructures. As an intermediary step, we prove that 2-WL and 2-IGNs are equivalent in distinguishing non-isomorphic graphs, partly answering an open problem raised in Maron et al. (2019). We also prove positive results for k-WL and k-IGNs as well as negative results for k-WL with a finite number of iterations. We then conduct experiments that support the theoretical results for MPNNs and 2-IGNs. Moreover, motivated by substructure counting and inspired by Murphy et al. (2019), we propose the Local Relational Pooling model and demonstrate that it is not only effective for substructure counting but also able to achieve competitive performance on molecular prediction tasks.

cs.LG↗