arXiv subjects

Kai Xu

Publications and source records attributed to Kai Xu.

At least 343 records · Page 19Linked to original sources

arXiv2020

Telescoping Density-Ratio Estimation

Density-ratio estimation via classification is a cornerstone of unsupervised learning. It has provided the foundation for state-of-the-art methods in representation learning and generative modelling, with the number of use-cases continuing to proliferate. However, it suffers from a critical limitation: it fails to accurately estimate ratios p/q for which the two densities differ significantly. Empirically, we find this occurs whenever the KL divergence between p and q exceeds tens of nats. To resolve this limitation, we introduce a new framework, telescoping density-ratio estimation (TRE), that enables the estimation of ratios between highly dissimilar densities in high-dimensional spaces. Our experiments demonstrate that TRE can yield substantial improvements over existing single-ratio methods for mutual information estimation, representation learning and energy-based modelling.

stat.ML↗

arXiv2020

WarpCore: A Library for fast Hash Tables on GPUs

Hash tables are ubiquitous. Properties such as an amortized constant time complexity for insertion and querying as well as a compact memory layout make them versatile associative data structures with manifold applications. The rapidly growing amount of data emerging in many fields motivated the need for accelerated hash tables designed for modern parallel architectures. In this work, we exploit the fast memory interface of modern GPUs together with a parallel hashing scheme tailored to improve global memory access patterns, to design WarpCore -- a versatile library of hash table data structures. Unique device-sided operations allow for building high performance data processing pipelines entirely on the GPU. Our implementation achieves up to 1.6 billion inserts and up to 4.3 billion retrievals per second on a single GV100 GPU thereby outperforming the state-of-the-art solutions cuDPP, SlabHash, and NVIDIA RAPIDS cuDF. This performance advantage becomes even more pronounced for high load factors of over $90\%$. To overcome the memory limitation of a single GPU, we scale our approach over a dense NVLink topology which gives us close-to-optimal weak scaling on DGX servers. We further show how WarpCore can be used for accelerating a real world bioinformatics application (metagenomic classification) with speedups of over two orders-of-magnitude against state-of-the-art CPU-based solutions. WC is written in C++/CUDA-C and is openly available at https://github.com/sleeepyjack/warpcore.

cs.DC↗

arXiv2020

Geometry of Localized Effective Theories, Exact Semi-classical Approximation and the Algebraic Index

In this paper we propose a general framework to study the quantum geometry of $σ$-models when they are effectively localized to small quantum fluctuations around constant maps. Such effective theories have surprising exact descriptions at all loops in terms of target geometry and can be rigorously formulated. We illustrate how to turn the physics idea of exact semi-classical approximation into a geometric set-up in this framework, using Gauss-Manin connection. As an application, we carry out this program in details by the example of topological quantum mechanics, and explain how to implement the idea of exact semi-classical approximation into a proof of the algebraic index theorem. The proof resembles much of the physics derivation of Atiyah-Singer index theorem and clarifies the geometric face of many other mathematical constructions.

math.QA↗

arXiv2020

PIE-NET: Parametric Inference of Point Cloud Edges

We introduce an end-to-end learnable technique to robustly identify feature edges in 3D point cloud data. We represent these edges as a collection of parametric curves (i.e.,lines, circles, and B-splines). Accordingly, our deep neural network, coined PIE-NET, is trained for parametric inference of edges. The network relies on a "region proposal" architecture, where a first module proposes an over-complete collection of edge and corner points, and a second module ranks each proposal to decide whether it should be considered. We train and evaluate our method on the ABC dataset, a large dataset of CAD models, and compare our results to those produced by traditional (non-learning) processing pipelines, as well as a recent deep learning based edge detector (EC-NET). Our results significantly improve over the state-of-the-art from both a quantitative and qualitative standpoint.

cs.CV↗

arXiv2020

AReLU: Attention-based Rectified Linear Unit

Element-wise activation functions play a critical role in deep neural networks via affecting the expressivity power and the learning dynamics. Learning-based activation functions have recently gained increasing attention and success. We propose a new perspective of learnable activation function through formulating them with element-wise attention mechanism. In each network layer, we devise an attention module which learns an element-wise, sign-based attention map for the pre-activation feature map. The attention map scales an element based on its sign. Adding the attention module with a rectified linear unit (ReLU) results in an amplification of positive elements and a suppression of negative ones, both with learned, data-adaptive parameters. We coin the resulting activation function Attention-based Rectified Linear Unit (AReLU). The attention module essentially learns an element-wise residue of the activated part of the input, as ReLU can be viewed as an identity transformation. This makes the network training more resistant to gradient vanishing. The learned attentive activation leads to well-focused activation of relevant regions of a feature map. Through extensive evaluations, we show that AReLU significantly boosts the performance of most mainstream network architectures with only two extra learnable parameters per layer introduced. Notably, AReLU facilitates fast network training under small learning rates, which makes it especially suited in the case of transfer learning and meta learning. Our source code has been released (see https://github.com/densechen/AReLU).

cs.LG↗

arXiv2020

Realizing a quantum generative adversarial network using a programmable superconducting processor

Generative adversarial networks are an emerging technique with wide applications in machine learning, which have achieved dramatic success in a number of challenging tasks including image and video generation. When equipped with quantum processors, their quantum counterparts--called quantum generative adversarial networks (QGANs)--may even exhibit exponential advantages in certain machine learning applications. Here, we report an experimental implementation of a QGAN using a programmable superconducting processor, in which both the generator and the discriminator are parameterized via layers of single- and multi-qubit quantum gates. The programmed QGAN runs automatically several rounds of adversarial learning with quantum gradients to achieve a Nash equilibrium point, where the generator can replicate data samples that mimic the ones from the training set. Our implementation is promising to scale up to noisy intermediate-scale quantum devices, thus paving the way for experimental explorations of quantum advantages in practical applications with near-term quantum technologies.

quant-ph↗

arXiv2020

SymmetryNet: Learning to Predict Reflectional and Rotational Symmetries of 3D Shapes from Single-View RGB-D Images

We study the problem of symmetry detection of 3D shapes from single-view RGB-D images, where severely missing data renders geometric detection approach infeasible. We propose an end-to-end deep neural network which is able to predict both reflectional and rotational symmetries of 3D objects present in the input RGB-D image. Directly training a deep model for symmetry prediction, however, can quickly run into the issue of overfitting. We adopt a multi-task learning approach. Aside from symmetry axis prediction, our network is also trained to predict symmetry correspondences. In particular, given the 3D points present in the RGB-D image, our network outputs for each 3D point its symmetric counterpart corresponding to a specific predicted symmetry. In addition, our network is able to detect for a given shape multiple symmetries of different types. We also contribute a benchmark of 3D symmetry detection based on single-view RGB-D images. Extensive evaluation on the benchmark demonstrates the strong generalization ability of our method, in terms of high accuracy of both symmetry axis prediction and counterpart estimation. In particular, our method is robust in handling unseen object instances with large variation in shape, multi-symmetry composition, as well as novel object categories.

cs.CV↗

arXiv2020

Generating Grasp Poses for a High-DOF Gripper Using Neural Networks

We present a learning-based method for representing grasp poses of a high-DOF hand using neural networks. Due to redundancy in such high-DOF grippers, there exists a large number of equally effective grasp poses for a given target object, making it difficult for the neural network to find consistent grasp poses. We resolve this ambiguity by generating an augmented dataset that covers many possible grasps for each target object and train our neural networks using a consistency loss function to identify a one-to-one mapping from objects to grasp poses. We further enhance the quality of neural-network-predicted grasp poses using a collision loss function to avoid penetrations. We use an object dataset that combines the BigBIRD Database, the KIT Database, the YCB Database, and the Grasp Dataset to show that our method can generate high-DOF grasp poses with higher accuracy than supervised learning baselines. The quality of the grasp poses is on par with the groundtruth poses in the dataset. In addition, our method is robust and can handle noisy object models such as those constructed from multi-view depth images, allowing our method to be implemented on a 25-DOF Shadow Hand hardware platform.

cs.RO↗

arXiv2020

Deep Differentiable Grasp Planner for High-DOF Grippers

We present an end-to-end algorithm for training deep neural networks to grasp novel objects. Our algorithm builds all the essential components of a grasping system using a forward-backward automatic differentiation approach, including the forward kinematics of the gripper, the collision between the gripper and the target object, and the metric for grasp poses. In particular, we show that a generalized Q1 grasp metric is defined and differentiable for inexact grasps generated by a neural network, and the derivatives of our generalized Q1 metric can be computed from a sensitivity analysis of the induced optimization problem. We show that the derivatives of the (self-)collision terms can be efficiently computed from a watertight triangle mesh of low-quality. Altogether, our algorithm allows for the computation of grasp poses for high-DOF grippers in an unsupervised mode with no ground truth data, or it improves the results in a supervised mode using a small dataset. Our new learning algorithm significantly simplifies the data preparation for learning-based grasping systems and leads to higher qualities of learned grasps on common 3D shape datasets [7, 49, 26, 25], achieving a 22% higher success rate on physical hardware and a 0.12 higher value on the Q1 grasp quality metric.

cs.RO↗

arXiv2020

Pentaquark components in low-lying baryon resonances

We study pentaquark states of both light $q^4\bar q$ and hidden heavy $q^3 Q\bar Q$ (q = u,d,s quark in SU(3) flavor symmetry; Q = c, b quark) systems with a general group theory approach in the constituent quark model, and the spectrum of light baryon resonances in the ansatz that the $l=1$ baryon states may consist of the $q^3$ as well as $q^4\bar q$ pentaquark component. The model is fitted to ground state baryons and light baryon resonances which are believed to be normal three-quark states. The work reveals that the $N(1535)1/2^{-}$ and $N(1520)3/2^-$ may consist of a large $q^4\bar q$ component while the $N(1895)1/2^{-}$ and $N(1875)3/2^-$ are respectively their partners, and the $N^+(1685)$ might be a $q^4\bar q$ state. By the way, a new set of color-spin-flavor-spatial wave function for $q^3 Q\bar Q$ systems in the compact pentaquark picture are constructed systematically for studying hidden charm pentaquark states.

hep-ph↗

arXiv2020

PQ-NET: A Generative Part Seq2Seq Network for 3D Shapes

We introduce PQ-NET, a deep neural network which represents and generates 3D shapes via sequential part assembly. The input to our network is a 3D shape segmented into parts, where each part is first encoded into a feature representation using a part autoencoder. The core component of PQ-NET is a sequence-to-sequence or Seq2Seq autoencoder which encodes a sequence of part features into a latent vector of fixed size, and the decoder reconstructs the 3D shape, one part at a time, resulting in a sequential assembly. The latent space formed by the Seq2Seq encoder encodes both part structure and fine part geometry. The decoder can be adapted to perform several generative tasks including shape autoencoding, interpolation, novel shape generation, and single-view 3D reconstruction, where the generated shapes are all composed of meaningful parts.

cs.CV↗

arXiv2020

MLCVNet: Multi-Level Context VoteNet for 3D Object Detection

In this paper, we address the 3D object detection task by capturing multi-level contextual information with the self-attention mechanism and multi-scale feature fusion. Most existing 3D object detection methods recognize objects individually, without giving any consideration on contextual information between these objects. Comparatively, we propose Multi-Level Context VoteNet (MLCVNet) to recognize 3D objects correlatively, building on the state-of-the-art VoteNet. We introduce three context modules into the voting and classifying stages of VoteNet to encode contextual information at different levels. Specifically, a Patch-to-Patch Context (PPC) module is employed to capture contextual information between the point patches, before voting for their corresponding object centroid points. Subsequently, an Object-to-Object Context (OOC) module is incorporated before the proposal and classification stage, to capture the contextual information between object candidates. Finally, a Global Scene Context (GSC) module is designed to learn the global scene context. We demonstrate these by capturing contextual information at patch, object and scene levels. Our method is an effective way to promote detection accuracy, achieving new state-of-the-art detection performance on challenging 3D object detection datasets, i.e., SUN RGBD and ScanNet. We also release our code at https://github.com/NUAAXQ/MLCVNet.

cs.CV↗

arXiv2020

Learning in the Frequency Domain

Deep neural networks have achieved remarkable success in computer vision tasks. Existing neural networks mainly operate in the spatial domain with fixed input sizes. For practical applications, images are usually large and have to be downsampled to the predetermined input size of neural networks. Even though the downsampling operations reduce computation and the required communication bandwidth, it removes both redundant and salient information obliviously, which results in accuracy degradation. Inspired by digital signal processing theories, we analyze the spectral bias from the frequency perspective and propose a learning-based frequency selection method to identify the trivial frequency components which can be removed without accuracy loss. The proposed method of learning in the frequency domain leverages identical structures of the well-known neural networks, such as ResNet-50, MobileNetV2, and Mask R-CNN, while accepting the frequency-domain information as the input. Experiment results show that learning in the frequency domain with static channel selection can achieve higher accuracy than the conventional spatial downsampling approach and meanwhile further reduce the input data size. Specifically for ImageNet classification with the same input size, the proposed method achieves 1.41% and 0.66% top-1 accuracy improvements on ResNet-50 and MobileNetV2, respectively. Even with half input size, the proposed method still improves the top-1 accuracy on ResNet-50 by 1%. In addition, we observe a 0.8% average precision improvement on Mask R-CNN for instance segmentation on the COCO dataset.

cs.CV↗

arXiv2020

Finite-system Multicriticality at the Superradiant Quantum Phase Transition

We demonstrate the existence of finite-system multicriticality in a qubit-boson model where biased qubits collectively coupled to a single-mode bosonic field. The interplay between biases and boson-qubit coupling produces a rich phase diagram showing multiple superradiant phases and phase boundaries of different orders. In particular, multiple phases can become indistinguishable in appropriate bias configurations, which is the signature of multicriticality. A series of universality classes characterizing these multicritical points are identified. Moreover, we present a trapped-ion realization with the potential to explore the multicritical phenomena experimentally using a small number of ions. The results open a novel way to probe multicritical universality classes in experiments.

quant-ph↗

arXiv2020

Generative Ratio Matching Networks

Deep generative models can learn to generate realistic-looking images, but many of the most effective methods are adversarial and involve a saddlepoint optimization, which requires a careful balancing of training between a generator network and a critic network. Maximum mean discrepancy networks (MMD-nets) avoid this issue by using kernel as a fixed adversary, but unfortunately, they have not on their own been able to match the generative quality of adversarial training. In this work, we take their insight of using kernels as fixed adversaries further and present a novel method for training deep generative models that does not involve saddlepoint optimization. We call our method generative ratio matching or GRAM for short. In GRAM, the generator and the critic networks do not play a zero-sum game against each other, instead, they do so against a fixed kernel. Thus GRAM networks are not only stable to train like MMD-nets but they also match and beat the generative quality of adversarially trained generative networks.

stat.ML↗

arXiv2020

DynamicPPL: Stan-like Speed for Dynamic Probabilistic Models

We present the preliminary high-level design and features of DynamicPPL.jl, a modular library providing a lightning-fast infrastructure for probabilistic programming. Besides a computational performance that is often close to or better than Stan, DynamicPPL provides an intuitive DSL that allows the rapid development of complex dynamic probabilistic programs. Being entirely written in Julia, a high-level dynamic programming language for numerical computing, DynamicPPL inherits a rich set of features available through the Julia ecosystem. Since DynamicPPL is a modular, stand-alone library, any probabilistic programming system written in Julia, such as Turing.jl, can use DynamicPPL to specify models and trace their model parameters. The main features of DynamicPPL are: 1) a meta-programming based DSL for specifying dynamic models using an intuitive tilde-based notation; 2) a tracing data-structure for tracking RVs in dynamic probabilistic models; 3) a rich contextual dispatch system allowing tailored behaviour during model execution; and 4) a user-friendly syntax for probabilistic queries. Finally, we show in a variety of experiments that DynamicPPL, in combination with Turing.jl, achieves computational performance that is often close to or better than Stan.

cs.LG↗

arXiv2020

Learning Part Generation and Assembly for Structure-aware Shape Synthesis

Learning powerful deep generative models for 3D shape synthesis is largely hindered by the difficulty in ensuring plausibility encompassing correct topology and reasonable geometry. Indeed, learning the distribution of plausible 3D shapes seems a daunting task for the holistic approaches, given the significant topological variations of 3D objects even within the same category. Enlightened by the fact that 3D shape structure is characterized as part composition and placement, we propose to model 3D shape variations with a part-aware deep generative network, coined as PAGENet. The network is composed of an array of per-part VAE-GANs, generating semantic parts composing a complete shape, followed by a part assembly module that estimates a transformation for each part to correlate and assemble them into a plausible structure. Through delegating the learning of part composition and part placement into separate networks, the difficulty of modeling structural variations of 3D shapes is greatly reduced. We demonstrate through both qualitative and quantitative evaluations that PAGENet generates 3D shapes with plausible, diverse and detailed structure, and show two applications, i.e., semantic shape segmentation and part-based shape editing.

cs.CV↗

arXiv2020

Higher-Rank Tensor Field Theory of Non-Abelian Fracton and Embeddon

We formulate a new class of tensor gauge field theories in any dimension that is a hybrid class between symmetric higher-rank tensor gauge theory (i.e., higher-spin gauge theory) and anti-symmetric tensor topological field theory. Our theory describes a mixed unitary phase interplaying between gapless and gapped topological order phases (which can live with or without Euclidean, Poincaré or anisotropic symmetry, at least in ultraviolet high or intermediate energy field theory, but not yet to a lattice cutoff scale). The "gauge structure" can be compact, continuous, abelian or non-abelian. Our theory sits outside the paradigm of Maxwell electromagnetic theory in 1865 and Yang-Mills isospin/color theory in 1954. We discuss its local gauge transformation in terms of the ungauged vector-like or tensor-like higher-moment global symmetry. The non-abelian gauge structure is caused by gauging the non-commutative symmetries: a higher-moment symmetry and a charge conjugation (particle-hole) symmetry. Vector global symmetries along time direction may exhibit time crystals. We explore the relation of these long-range entangled matters to a non-abelian generalization of Fracton order in condensed matter, a field theory formulation of foliation, the spacetime embedding and Embeddon that we newly introduce, and possible fundamental physics applications to dark matter or dark energy.

hep-th↗