arXiv subjects

Hongkai Xiong

Publications and source records attributed to Hongkai Xiong.

At least 73 records · Page 4Linked to original sources

arXiv2022

All-optical graph representation learning using integrated diffractive photonic computing units

Photonic neural networks perform brain-inspired computations using photons instead of electrons that can achieve substantially improved computing performance. However, existing architectures can only handle data with regular structures, e.g., images or videos, but fail to generalize to graph-structured data beyond Euclidean space, e.g., social networks or document co-citation networks. Here, we propose an all-optical graph representation learning architecture, termed diffractive graph neural network (DGNN), based on the integrated diffractive photonic computing units (DPUs) to address this limitation. Specifically, DGNN optically encodes node attributes into strip optical waveguides, which are transformed by DPUs and aggregated by on-chip optical couplers to extract their feature representations. Each DPU comprises successive passive layers of metalines to modulate the electromagnetic optical field via diffraction, where the metaline structures are learnable parameters shared across graph nodes. DGNN captures complex dependencies among the node neighborhoods and eliminates the nonlinear transition functions during the light-speed optical message passing over graph structures. We demonstrate the use of DGNN extracted features for node and graph-level classification tasks with benchmark databases and achieve superior performance. Our work opens up a new direction for designing application-specific integrated photonic circuits for high-efficiency processing of large-scale graph data structures using deep learning.

cs.ET↗

arXiv2022

Bag of Instances Aggregation Boosts Self-supervised Distillation

Recent advances in self-supervised learning have experienced remarkable progress, especially for contrastive learning based methods, which regard each image as well as its augmentations as an individual class and try to distinguish them from all other images. However, due to the large quantity of exemplars, this kind of pretext task intrinsically suffers from slow convergence and is hard for optimization. This is especially true for small-scale models, in which we find the performance drops dramatically comparing with its supervised counterpart. In this paper, we propose a simple but effective distillation strategy for unsupervised learning. The highlight is that the relationship among similar samples counts and can be seamlessly transferred to the student to boost the performance. Our method, termed as BINGO, which is short for Bag of InstaNces aGgregatiOn, targets at transferring the relationship learned by the teacher to the student. Here bag of instances indicates a set of similar samples constructed by the teacher and are grouped within a bag, and the goal of distillation is to aggregate compact representations over the student with respect to instances in a bag. Notably, BINGO achieves new state-of-the-art performance on small-scale models, i.e., 65.5% and 68.9% top-1 accuracies with linear evaluation on ImageNet, using ResNet-18 and ResNet-34 as the backbones respectively, surpassing baselines (52.5% and 57.4% top-1 accuracies) by a significant margin. The code is available at https://github.com/haohang96/bingo.

cs.CV↗

arXiv2022

Motion-aware Contrastive Video Representation Learning via Foreground-background Merging

In light of the success of contrastive learning in the image domain, current self-supervised video representation learning methods usually employ contrastive loss to facilitate video representation learning. When naively pulling two augmented views of a video closer, the model however tends to learn the common static background as a shortcut but fails to capture the motion information, a phenomenon dubbed as background bias. Such bias makes the model suffer from weak generalization ability, leading to worse performance on downstream tasks such as action recognition. To alleviate such bias, we propose \textbf{F}oreground-b\textbf{a}ckground \textbf{Me}rging (FAME) to deliberately compose the moving foreground region of the selected video onto the static background of others. Specifically, without any off-the-shelf detector, we extract the moving foreground out of background regions via the frame difference and color statistics, and shuffle the background regions among the videos. By leveraging the semantic consistency between the original clips and the fused ones, the model focuses more on the motion patterns and is debiased from the background shortcut. Extensive experiments demonstrate that FAME can effectively resist background cheating and thus achieve the state-of-the-art performance on downstream tasks across UCF101, HMDB51, and Diving48 datasets. The code and configurations are released at https://github.com/Mark12Ding/FAME.

cs.CV↗

arXiv2022

Graph Neural Networks With Lifting-based Adaptive Graph Wavelets

Spectral-based graph neural networks (SGNNs) have been attracting increasing attention in graph representation learning. However, existing SGNNs are limited in implementing graph filters with rigid transforms (e.g., graph Fourier or predefined graph wavelet transforms) and cannot adapt to signals residing on graphs and tasks at hand. In this paper, we propose a novel class of graph neural networks that realizes graph filters with adaptive graph wavelets. Specifically, the adaptive graph wavelets are learned with neural network-parameterized lifting structures, where structure-aware attention-based lifting operations (i.e., prediction and update operations) are developed to jointly consider graph structures and node features. We propose to lift based on diffusion wavelets to alleviate the structural information loss induced by partitioning non-bipartite graphs. By design, the locality and sparsity of the resulting wavelet transform as well as the scalability of the lifting structure are guaranteed. We further derive a soft-thresholding filtering operation by learning sparse graph representations in terms of the learned wavelets, yielding a localized, efficient, and scalable wavelet-based graph filters. To ensure that the learned graph representations are invariant to node permutations, a layer is employed at the input of the networks to reorder the nodes according to their local topology information. We evaluate the proposed networks in both node-level and graph-level representation learning tasks on benchmark citation and bioinformatics graph datasets. Extensive experiments demonstrate the superiority of the proposed networks over existing SGNNs in terms of accuracy, efficiency, and scalability.

cs.LG↗

arXiv2021

Center-wise Local Image Mixture For Contrastive Representation Learning

Contrastive learning based on instance discrimination trains model to discriminate different transformations of the anchor sample from other samples, which does not consider the semantic similarity among samples. This paper proposes a new kind of contrastive learning method, named CLIM, which uses positives from other samples in the dataset. This is achieved by searching local similar samples of the anchor, and selecting samples that are closer to the corresponding cluster center, which we denote as center-wise local image selection. The selected samples are instantiated via an data mixture strategy, which performs as a smoothing regularization. As a result, CLIM encourages both local similarity and global aggregation in a robust way, which we find is beneficial for feature representation. Besides, we introduce \emph{multi-resolution} augmentation, which enables the representation to be scale invariant. We reach 75.5% top-1 accuracy with linear evaluation over ResNet-50, and 59.3% top-1 accuracy when fine-tuned with only 1% labels.

cs.CV↗

arXiv2021

Optimization-based Block Coordinate Gradient Coding

Existing gradient coding schemes introduce identical redundancy across the coordinates of gradients and hence cannot fully utilize the computation results from partial stragglers. This motivates the introduction of diverse redundancies across the coordinates of gradients. This paper considers a distributed computation system consisting of one master and $N$ workers characterized by a general partial straggler model and focuses on solving a general large-scale machine learning problem with $L$ model parameters. We show that it is sufficient to provide at most $N$ levels of redundancies for tolerating $0, 1,\cdots, N-1$ stragglers, respectively. Consequently, we propose an optimal block coordinate gradient coding scheme based on a stochastic optimization problem that optimizes the partition of the $L$ coordinates into $N$ blocks, each with identical redundancy, to minimize the expected overall runtime for collaboratively computing the gradient. We obtain an optimal solution using a stochastic projected subgradient method and propose two low-complexity approximate solutions with closed-from expressions, for the stochastic optimization problem. We also show that under a shifted-exponential distribution, for any $L$, the expected overall runtimes of the two approximate solutions and the minimum overall runtime have sub-linear multiplicative gaps in $N$. To the best of our knowledge, this is the first work that optimizes the redundancies of gradient coding introduced across the coordinates of gradients.

cs.DC↗

arXiv2021

Message Passing in Graph Convolution Networks via Adaptive Filter Banks

Graph convolution networks, like message passing graph convolution networks (MPGCNs), have been a powerful tool in representation learning of networked data. However, when data is heterogeneous, most architectures are limited as they employ a single strategy to handle multi-channel graph signals and they typically focus on low-frequency information. In this paper, we present a novel graph convolution operator, termed BankGCN, which keeps benefits of message passing models, but extends their capabilities beyond `low-pass' features. It decomposes multi-channel signals on graphs into subspaces and handles particular information in each subspace with an adapted filter. The filters of all subspaces have different frequency responses and together form a filter bank. Furthermore, each filter in the spectral domain corresponds to a message passing scheme, and diverse schemes are implemented via the filter bank. Importantly, the filter bank and the signal decomposition are jointly learned to adapt to the spectral characteristics of data and to target applications. Furthermore, this is implemented almost without extra parameters in comparison with most existing MPGCNs. Experimental results show that the proposed convolution operator permits to achieve excellent performance in graph classification on a collection of benchmark graph datasets.

cs.LG↗

arXiv2021

Multi-dataset Pretraining: A Unified Model for Semantic Segmentation

Collecting annotated data for semantic segmentation is time-consuming and hard to scale up. In this paper, we for the first time propose a unified framework, termed as Multi-Dataset Pretraining, to take full advantage of the fragmented annotations of different datasets. The highlight is that the annotations from different domains can be efficiently reused and consistently boost performance for each specific domain. This is achieved by first pretraining the network via the proposed pixel-to-prototype contrastive loss over multiple datasets regardless of their taxonomy labels, and followed by fine-tuning the pretrained model over specific dataset as usual. In order to better model the relationship among images and classes from different datasets, we extend the pixel level embeddings via cross dataset mixing and propose a pixel-to-class sparse coding strategy that explicitly models the pixel-class similarity over the manifold embedding space. In this way, we are able to increase intra-class compactness and inter-class separability, as well as considering inter-class similarity across different datasets for better transferability. Experiments conducted on several benchmarks demonstrate its superior performance. Notably, MDP consistently outperforms the pretrained models over ImageNet by a considerable margin, while only using less than 10% samples for pretraining.

cs.CV↗

arXiv2021

Seed the Views: Hierarchical Semantic Alignment for Contrastive Representation Learning

Self-supervised learning based on instance discrimination has shown remarkable progress. In particular, contrastive learning, which regards each image as well as its augmentations as an individual class and tries to distinguish them from all other images, has been verified effective for representation learning. However, pushing away two images that are de facto similar is suboptimal for general representation. In this paper, we propose a hierarchical semantic alignment strategy via expanding the views generated by a single image to \textbf{Cross-samples and Multi-level} representation, and models the invariance to semantically similar images in a hierarchical way. This is achieved by extending the contrastive loss to allow for multiple positives per anchor, and explicitly pulling semantically similar images/patches together at different layers of the network. Our method, termed as CsMl, has the ability to integrate multi-level visual representations across samples in a robust way. CsMl is applicable to current contrastive learning based methods and consistently improves the performance. Notably, using the moco as an instantiation, CsMl achieves a \textbf{76.6\% }top-1 accuracy with linear evaluation using ResNet-50 as backbone, and \textbf{66.7\%} and \textbf{75.1\%} top-1 accuracy with only 1\% and 10\% labels, respectively. \textbf{All these numbers set the new state-of-the-art.}

cs.CV↗

arXiv2021

Spatial-Temporal Transformer Networks for Traffic Flow Forecasting

Traffic forecasting has emerged as a core component of intelligent transportation systems. However, timely accurate traffic forecasting, especially long-term forecasting, still remains an open challenge due to the highly nonlinear and dynamic spatial-temporal dependencies of traffic flows. In this paper, we propose a novel paradigm of Spatial-Temporal Transformer Networks (STTNs) that leverages dynamical directed spatial dependencies and long-range temporal dependencies to improve the accuracy of long-term traffic forecasting. Specifically, we present a new variant of graph neural networks, named spatial transformer, by dynamically modeling directed spatial dependencies with self-attention mechanism to capture realtime traffic conditions as well as the directionality of traffic flows. Furthermore, different spatial dependency patterns can be jointly modeled with multi-heads attention mechanism to consider diverse relationships related to different factors (e.g. similarity, connectivity and covariance). On the other hand, the temporal transformer is utilized to model long-range bidirectional temporal dependencies across multiple time steps. Finally, they are composed as a block to jointly model the spatial-temporal dependencies for accurate traffic prediction. Compared to existing works, the proposed model enables fast and scalable training over a long range spatial-temporal dependencies. Experiment results demonstrate that the proposed model achieves competitive results compared with the state-of-the-arts, especially forecasting long-term traffic flows on real-world PeMS-Bay and PeMSD7(M) datasets.

eess.SP↗

arXiv2021

K-Shot Contrastive Learning of Visual Features with Multiple Instance Augmentations

In this paper, we propose the $K$-Shot Contrastive Learning (KSCL) of visual features by applying multiple augmentations to investigate the sample variations within individual instances. It aims to combine the advantages of inter-instance discrimination by learning discriminative features to distinguish between different instances, as well as intra-instance variations by matching queries against the variants of augmented samples over instances. Particularly, for each instance, it constructs an instance subspace to model the configuration of how the significant factors of variations in $K$-shot augmentations can be combined to form the variants of augmentations. Given a query, the most relevant variant of instances is then retrieved by projecting the query onto their subspaces to predict the positive instance class. This generalizes the existing contrastive learning that can be viewed as a special one-shot case. An eigenvalue decomposition is performed to configure instance subspaces, and the embedding network can be trained end-to-end through the differentiable subspace configuration. Experiment results demonstrate the proposed $K$-shot contrastive learning achieves superior performances to the state-of-the-art unsupervised methods.

cs.CV↗

arXiv2020

Batch Normalization with Enhanced Linear Transformation

Batch normalization (BN) is a fundamental unit in modern deep networks, in which a linear transformation module was designed for improving BN's flexibility of fitting complex data distributions. In this paper, we demonstrate properly enhancing this linear transformation module can effectively improve the ability of BN. Specifically, rather than using a single neuron, we propose to additionally consider each neuron's neighborhood for calculating the outputs of the linear transformation. Our method, named BNET, can be implemented with 2-3 lines of code in most deep learning libraries. Despite the simplicity, BNET brings consistent performance gains over a wide range of backbones and visual benchmarks. Moreover, we verify that BNET accelerates the convergence of network training and enhances spatial information by assigning the important neurons with larger weights accordingly. The code is available at https://github.com/yuhuixu1993/BNET.

cs.CV↗

arXiv2020

Key-Point Sequence Lossless Compression for Intelligent Video Analysis

Feature coding has been recently considered to facilitate intelligent video analysis for urban computing. Instead of raw videos, extracted features in the front-end are encoded and transmitted to the back-end for further processing. In this article, we present a lossless key-point sequence compression approach for efficient feature coding. The essence of this predict-and-encode strategy is to eliminate the spatial and temporal redundancies of key points in videos. Multiple prediction modes with an adaptive mode selection method are proposed to handle key-point sequences with various structures and motion. Experimental results validate the effectiveness of the proposed scheme on four types of widely used key-point sequences in video analysis.

cs.MM↗

arXiv2020

Attribute Mix: Semantic Data Augmentation for Fine Grained Recognition

Collecting fine-grained labels usually requires expert-level domain knowledge and is prohibitive to scale up. In this paper, we propose Attribute Mix, a data augmentation strategy at attribute level to expand the fine-grained samples. The principle lies in that attribute features are shared among fine-grained sub-categories, and can be seamlessly transferred among images. Toward this goal, we propose an automatic attribute mining approach to discover attributes that belong to the same super-category, and Attribute Mix is operated by mixing semantically meaningful attribute features from two images. Attribute Mix is a simple but effective data augmentation strategy that can significantly improve the recognition performance without increasing the inference budgets. Furthermore, since attributes can be shared among images from the same super-category, we further enrich the training samples with attribute level labels using images from the generic domain. Experiments on widely used fine-grained benchmarks demonstrate the effectiveness of our proposed method.

cs.CV↗

arXiv2020

Distilling Object Detectors with Task Adaptive Regularization

Current state-of-the-art object detectors are at the expense of high computational costs and are hard to deploy to low-end devices. Knowledge distillation, which aims at training a smaller student network by transferring knowledge from a larger teacher model, is one of the promising solutions for model miniaturization. In this paper, we investigate each module of a typical detector in depth, and propose a general distillation framework that adaptively transfers knowledge from teacher to student according to the task specific priors. The intuition is that simply distilling all information from teacher to student is not advisable, instead we should only borrow priors from the teacher model where the student cannot perform well. Towards this goal, we propose a region proposal sharing mechanism to interflow region responses between the teacher and student models. Based on this, we adaptively transfer knowledge at three levels, \emph{i.e.}, feature backbone, classification head, and bounding box regression head, according to which model performs more reasonably. Furthermore, considering that it would introduce optimization dilemma when minimizing distillation loss and detection loss simultaneously, we propose a distillation decay strategy to help improve model generalization via gradually reducing the distillation penalty. Experiments on widely used detection benchmarks demonstrate the effectiveness of our method. In particular, using Faster R-CNN with FPN as an instantiation, we achieve an accuracy of $39.0\%$ with Resnet-50 on COCO dataset, which surpasses the baseline $36.3\%$ by $2.7\%$ points, and even better than the teacher model with $38.5\%$ mAP.

cs.CV↗

arXiv2020

Graph Pooling with Node Proximity for Hierarchical Representation Learning

Graph neural networks have attracted wide attentions to enable representation learning of graph data in recent works. In complement to graph convolution operators, graph pooling is crucial for extracting hierarchical representation of graph data. However, most recent graph pooling methods still fail to efficiently exploit the geometry of graph data. In this paper, we propose a novel graph pooling strategy that leverages node proximity to improve the hierarchical representation learning of graph data with their multi-hop topology. Node proximity is obtained by harmonizing the kernel representation of topology information and node features. Implicit structure-aware kernel representation of topology information allows efficient graph pooling without explicit eigendecomposition of the graph Laplacian. Similarities of node signals are adaptively evaluated with the combination of the affine transformation and kernel trick using the Gaussian RBF function. Experimental results demonstrate that the proposed graph pooling strategy is able to achieve state-of-the-art performance on a collection of public graph classification benchmark datasets.

cs.LG↗

arXiv2020

TRP: Trained Rank Pruning for Efficient Deep Neural Networks

To enable DNNs on edge devices like mobile phones, low-rank approximation has been widely adopted because of its solid theoretical rationale and efficient implementations. Several previous works attempted to directly approximate a pretrained model by low-rank decomposition; however, small approximation errors in parameters can ripple over a large prediction loss. As a result, performance usually drops significantly and a sophisticated effort on fine-tuning is required to recover accuracy. Apparently, it is not optimal to separate low-rank approximation from training. Unlike previous works, this paper integrates low rank approximation and regularization into the training process. We propose Trained Rank Pruning (TRP), which alternates between low rank approximation and training. TRP maintains the capacity of the original network while imposing low-rank constraints during training. A nuclear regularization optimized by stochastic sub-gradient descent is utilized to further promote low rank in TRP. The TRP trained network inherently has a low-rank structure, and is approximated with negligible performance loss, thus eliminating the fine-tuning process after low rank decomposition. The proposed method is comprehensively evaluated on CIFAR-10 and ImageNet, outperforming previous compression methods using low rank approximation.

cs.LG↗

arXiv2020

PC-DARTS: Partial Channel Connections for Memory-Efficient Architecture Search

Differentiable architecture search (DARTS) provided a fast solution in finding effective network architectures, but suffered from large memory and computing overheads in jointly training a super-network and searching for an optimal architecture. In this paper, we present a novel approach, namely, Partially-Connected DARTS, by sampling a small part of super-network to reduce the redundancy in exploring the network space, thereby performing a more efficient search without comprising the performance. In particular, we perform operation search in a subset of channels while bypassing the held out part in a shortcut. This strategy may suffer from an undesired inconsistency on selecting the edges of super-net caused by sampling different channels. We alleviate it using edge normalization, which adds a new set of edge-level parameters to reduce uncertainty in search. Thanks to the reduced memory cost, PC-DARTS can be trained with a larger batch size and, consequently, enjoys both faster speed and higher training stability. Experimental results demonstrate the effectiveness of the proposed method. Specifically, we achieve an error rate of 2.57% on CIFAR10 with merely 0.1 GPU-days for architecture search, and a state-of-the-art top-1 error rate of 24.2% on ImageNet (under the mobile setting) using 3.8 GPU-days for search. Our code has been made available at: https://github.com/yuhuixu1993/PC-DARTS.

cs.CV↗