1 code implementation • 12 Apr 2022 • Minbin Huang, Zhijian Huang, Changlin Li, Xin Chen, Hang Xu, Zhenguo Li, Xiaodan Liang
It is able to find top 0. 16\% and 0. 29\% architectures on average on two search spaces under the budget of only 50 models.
no code implementations • 9 Apr 2022 • Jianan Wang, Guansong Lu, Hang Xu, Zhenguo Li, Chunjing Xu, Yanwei Fu
Existing text-guided image manipulation methods aim to modify the appearance of the image or to edit a few objects in a virtual or simple scenario, which is far from practical application.
1 code implementation • ICLR 2022 • Shoukang Hu, Ruochen Wang, Lanqing Hong, Zhenguo Li, Cho-Jui Hsieh, Jiashi Feng
Efficient performance estimation of architectures drawn from large search spaces is essential to Neural Architecture Search.
2 code implementations • ICLR 2022 • Xingchen Wan, Binxin Ru, Pedro M. Esperança, Zhenguo Li
Searching for the architecture cells is a dominant paradigm in NAS.
no code implementations • 15 Mar 2022 • Kaican Li, Kai Chen, Haoyu Wang, Lanqing Hong, Chaoqiang Ye, Jianhua Han, Yukuai Chen, Wei zhang, Chunjing Xu, Dit-yan Yeung, Xiaodan Liang, Zhenguo Li, Hang Xu
One main reason that impedes the development of truly reliably self-driving systems is the lack of public datasets for evaluating the performance of object detectors on corner cases.
no code implementations • ICLR 2022 • Han Shi, Jiahui Gao, Hang Xu, Xiaodan Liang, Zhenguo Li, Lingpeng Kong, Stephen M. S. Lee, James T. Kwok
Recently over-smoothing phenomenon of Transformer-based models is observed in both vision and language fields.
1 code implementation • ICLR 2022 • Liyuan Wang, Xingxing Zhang, Kuo Yang, Longhui Yu, Chongxuan Li, Lanqing Hong, Shifeng Zhang, Zhenguo Li, Yi Zhong, Jun Zhu
In this work, we propose memory replay with data compression (MRDC) to reduce the storage cost of old training samples and thus increase their amount that can be stored in the memory buffer.
1 code implementation • 27 Jan 2022 • Weijun Hong, Guilin Li, Weinan Zhang, Ruiming Tang, Yunhe Wang, Zhenguo Li, Yong Yu
Neural architecture search (NAS) has shown encouraging results in automating the architecture design.
no code implementations • 13 Dec 2021 • Sarah Parisot, Pedro M. Esperanca, Steven McDonagh, Tamas J. Madarasz, Yongxin Yang, Zhenguo Li
In this work, we introduce a novel strategy for long-tail recognition that addresses the tail classes' few-shot problem via training-free knowledge transfer.
no code implementations • 10 Dec 2021 • Qi Sun, Hexin Dong, Zewei Chen, Jiacheng Sun, Zhenguo Li, Bin Dong
Gradient-based methods for the distributed training of residual networks (ResNets) typically require a forward pass of the input data, followed by back-propagating the error gradient to update model parameters, which becomes time-consuming as the network goes deeper.
no code implementations • 7 Dec 2021 • Tianyang Hu, Jun Wang, Wenjia Wang, Zhenguo Li
Comparing to cross-entropy, square loss has comparable generalization error but noticeable advantages in robustness and model calibration.
1 code implementation • NeurIPS 2021 • Hang Lai, Jian Shen, Weinan Zhang, Yimin Huang, Xing Zhang, Ruiming Tang, Yong Yu, Zhenguo Li
Model-based reinforcement learning has attracted wide attention due to its superior sample efficiency.
no code implementations • NeurIPS 2021 • Muhammad Awais, Fengwei Zhou, Chuanlong Xie, Jiawei Li, Sung-Ho Bae, Zhenguo Li
First, we theoretically show the transferability of robustness from an adversarially trained teacher model to a student model with the help of mixup augmentation.
no code implementations • ICLR 2022 • Lewei Yao, Runhui Huang, Lu Hou, Guansong Lu, Minzhe Niu, Hang Xu, Xiaodan Liang, Zhenguo Li, Xin Jiang, Chunjing Xu
In this paper, we introduce a large-scale Fine-grained Interactive Language-Image Pre-training (FILIP) to achieve finer-level alignment through a cross-modal late interaction mechanism, which uses a token-wise maximum similarity between visual and textual tokens to guide the contrastive objective.
1 code implementation • 5 Nov 2021 • Chenxu Zhu, Bo Chen, Weinan Zhang, Jincai Lai, Ruiming Tang, Xiuqiang He, Zhenguo Li, Yong Yu
To address these three issues mentioned above, we propose Automatic Interaction Machine (AIM) with three core components, namely, Feature Interaction Search (FIS), Interaction Function Search (IFS) and Embedding Dimension Search (EDS), to select significant feature interactions, appropriate interaction functions and necessary embedding dimensions automatically in a unified framework.
no code implementations • NeurIPS 2021 • Chen Zhang, Shifeng Zhang, Fabio Maria Carlucci, Zhenguo Li
To eliminate the requirement of saving separate models for different target datasets, we propose a novel setting that starts from a pretrained deep generative model and compresses the data batches while adapting the model with a dynamical system for only one epoch.
no code implementations • NeurIPS 2021 • Shifeng Zhang, Ning Kang, Tom Ryder, Zhenguo Li
In this paper, we discuss lossless compression using normalizing flows which have demonstrated a great capacity for achieving high compression ratios.
no code implementations • ICLR 2022 • Xiaojiang Yang, Yi Wang, Jiacheng Sun, Xing Zhang, Shifeng Zhang, Zhenguo Li, Junchi Yan
Nonlinear ICA is a fundamental problem in machine learning, aiming to identify the underlying independent components (sources) from data which is assumed to be a nonlinear function (mixing function) of these sources.
no code implementations • ICLR 2022 • Yao Zhu, Jiacheng Sun, Zhenguo Li
Adversarial transferability enables attackers to generate adversarial examples from the source model to attack the target model, which has raised security concerns about the deployment of DNNs in practice.
no code implementations • NeurIPS Workshop ImageNet_PPF 2021 • Dapeng Hu, Shipeng Yan, Qizhengqiu Lu, Lanqing Hong, Hailin Hu, Yifan Zhang, Zhenguo Li, Xinchao Wang, Jiashi Feng
Prior works on self-supervised pre-training focus on the joint training scenario, where massive unlabeled data are assumed to be given as input all at once, and only then is a learner trained.
no code implementations • NeurIPS Workshop DLDE 2021 • Qi Sun, Hexin Dong, Zewei Chen, Weizhen Dian, Jiacheng Sun, Yitong Sun, Zhenguo Li, Bin Dong
Backpropagation algorithm is indispensable for training modern residual networks (ResNets) and usually tends to be time-consuming due to its inherent algorithmic lockings.
no code implementations • 13 Sep 2021 • Kaichen Zhou, Lanqing Hong, Shoukang Hu, Fengwei Zhou, Binxin Ru, Jiashi Feng, Zhenguo Li
In view of these, we propose DHA, which achieves joint optimization of Data augmentation policy, Hyper-parameter and Architecture.
no code implementations • ICCV 2021 • Haoyue Bai, Fengwei Zhou, Lanqing Hong, Nanyang Ye, S. -H. Gary Chan, Zhenguo Li
In this work, we propose robust Neural Architecture Search for OoD generalization (NAS-OoD), which optimizes the architecture with respect to its performance on generated OoD data by gradient descent.
Ranked #1 on
Domain Generalization
on NICO Animal
no code implementations • ICCV 2021 • Muhammad Awais, Fengwei Zhou, Hang Xu, Lanqing Hong, Ping Luo, Sung-Ho Bae, Zhenguo Li
Extensive Unsupervised Domain Adaptation (UDA) studies have shown great success in practice by learning transferable representations across a labeled source domain and an unlabeled target domain with deep models.
1 code implementation • ICCV 2021 • Kai Chen, Lanqing Hong, Hang Xu, Zhenguo Li, Dit-yan Yeung
By pre-training on SODA10M, a large-scale autonomous driving dataset, MultiSiam exceeds the ImageNet pre-trained MoCo-v2, demonstrating the potential of domain-specific pre-training.
no code implementations • ICCV 2021 • Yao Zhu, Jiacheng Ma, Jiacheng Sun, Zewei Chen, Rongxin Jiang, Zhenguo Li
We find that adversarial training contributes to obtaining an energy function that is flat and has low energy around the real data, which is the key for generative capability.
no code implementations • ICCV 2021 • Lewei Yao, Renjie Pi, Hang Xu, Wei zhang, Zhenguo Li, Tong Zhang
In this paper, we investigate the knowledge distillation (KD) strategy for object detection and propose an effective framework applicable to both homogeneous and heterogeneous student-teacher pairs.
no code implementations • ICCV 2021 • Hang Xu, Ning Kang, Gengwei Zhang, Chuanlong Xie, Xiaodan Liang, Zhenguo Li
Fine-tuning from pre-trained ImageNet models has been a simple, effective, and popular approach for various computer vision tasks.
no code implementations • 15 Jul 2021 • Jiahui Gao, Hang Xu, Han Shi, Xiaozhe Ren, Philip L. H. Yu, Xiaodan Liang, Xin Jiang, Zhenguo Li
Transformer-based pre-trained language models like BERT and its variants have recently achieved promising performance in various natural language processing (NLP) tasks.
no code implementations • 21 Jun 2021 • Jianhua Han, Xiwen Liang, Hang Xu, Kai Chen, Lanqing Hong, Jiageng Mao, Chaoqiang Ye, Wei zhang, Zhenguo Li, Xiaodan Liang, Chunjing Xu
Experiments show that SODA10M can serve as a promising pre-training dataset for different self-supervised learning methods, which gives superior performance when fine-tuning with different downstream tasks (i. e., detection, semantic/instance segmentation) in autonomous driving domain.
1 code implementation • 21 Jun 2021 • Jiageng Mao, Minzhe Niu, Chenhan Jiang, Hanxue Liang, Jingheng Chen, Xiaodan Liang, Yamin Li, Chaoqiang Ye, Wei zhang, Zhenguo Li, Jie Yu, Hang Xu, Chunjing Xu
To facilitate future research on exploiting unlabeled data for 3D detection, we additionally provide a benchmark in which we reproduce and evaluate a variety of self-supervised and semi-supervised methods on the ONCE dataset.
no code implementations • CVPR 2021 • Aoxue Li, Zhenguo Li
To this end, we propose a simple yet effective Transformation Invariant Principle (TIP) that can be flexibly applied to various meta-learning models for boosting the detection performance on novel class objects.
1 code implementation • CVPR 2021 • Nanyang Ye, Jingxuan Tang, Huayu Deng, Xiao-Yun Zhou, Qianxiao Li, Zhenguo Li, Guang-Zhong Yang, Zhanxing Zhu
To the best of our knowledge, this is one of the first to adopt differentiable environment splitting method to enable stable predictions across environments without environment index information, which achieves the state-of-the-art performance on datasets with strong spurious correlation, such as Colored MNIST.
no code implementations • 15 Jun 2021 • Han-Jia Ye, Da-Wei Zhou, Lanqing Hong, Zhenguo Li, Xiu-Shen Wei, De-Chuan Zhan
One single instance could possess multiple portraits and reveal diverse relationships with others according to different contexts.
no code implementations • NeurIPS 2021 • Haotian Ye, Chuanlong Xie, Tianle Cai, Ruichen Li, Zhenguo Li, LiWei Wang
We also introduce a new concept of expansion function, which characterizes to what extent the variance is amplified in the test domains over the training domains, and therefore give a quantitative meaning of invariant features.
no code implementations • CVPR 2021 • Lewei Yao, Renjie Pi, Hang Xu, Wei zhang, Zhenguo Li, Tong Zhang
For student morphism, weight inheritance strategy is adopted, allowing the student to flexibly update its architecture while fully utilize the predecessor's weights, which considerably accelerates the search; To facilitate dynamic distillation, an elastic teacher pool is trained via integrated progressive shrinking strategy, from which teacher detectors can be sampled without additional cost in subsequent searches.
1 code implementation • CVPR 2021 • Yawen Duan, Xin Chen, Hang Xu, Zewei Chen, Xiaodan Liang, Tong Zhang, Zhenguo Li
While existing NAS methods mostly design architectures on a single task, algorithms that look beyond single-task search are surging to pursue a more efficient and universal solution across various tasks.
no code implementations • 13 May 2021 • Wenqi Shao, Hang Yu, Zhaoyang Zhang, Hang Xu, Zhenguo Li, Ping Luo
To address this problem, we develop a probability-based pruning algorithm, called batch whitening channel pruning (BWCP), which can stochastically discard unimportant channels by modeling the probability of a channel being activated.
no code implementations • ICLR 2022 • Dapeng Hu, Shipeng Yan, Qizhengqiu Lu, Lanqing Hong, Hailin Hu, Yifan Zhang, Zhenguo Li, Xinchao Wang, Jiashi Feng
Prior works on self-supervised pre-training focus on the joint training scenario, where massive unlabeled data are assumed to be given as input all at once, and only then is a learner trained.
no code implementations • CVPR 2021 • Shifeng Zhang, Chen Zhang, Ning Kang, Zhenguo Li
We also propose a lossless compression algorithm based on iVPF.
no code implementations • 8 Mar 2021 • Jian Ding, Enze Xie, Hang Xu, Chenhan Jiang, Zhenguo Li, Ping Luo, Gui-Song Xia
Unsupervised pre-training aims at learning transferable features that are beneficial for downstream tasks.
1 code implementation • 25 Feb 2021 • Han Shi, Jiahui Gao, Xiaozhe Ren, Hang Xu, Xiaodan Liang, Zhenguo Li, James T. Kwok
A surprising result is that diagonal elements in the attention map are the least important compared with other attention positions.
1 code implementation • ICLR 2021 • Peidong Liu, Gengwei Zhang, Bochao Wang, Hang Xu, Xiaodan Liang, Yong Jiang, Zhenguo Li
For object detection, the well-established classification and regression loss functions have been carefully designed by considering diverse learning challenges.
2 code implementations • ICCV 2021 • Enze Xie, Jian Ding, Wenhai Wang, Xiaohang Zhan, Hang Xu, Peize Sun, Zhenguo Li, Ping Luo
Unlike most recent methods that focused on improving accuracy of image classification, we present a novel contrastive learning approach, named DetCo, which fully explores the contrasts between global image and local image patches to learn discriminative representations for object detection.
no code implementations • 21 Jan 2021 • Haotian Ye, Chuanlong Xie, Yue Liu, Zhenguo Li
One of the definitions of OOD accuracy is worst-domain accuracy.
no code implementations • 5 Jan 2021 • Qijun Luo, Zhili Liu, Lanqing Hong, Chongxuan Li, Kuo Yang, Liyuan Wang, Fengwei Zhou, Guilin Li, Zhenguo Li, Jun Zhu
Semi-supervised domain adaptation (SSDA), which aims to learn models in a partially labeled target domain with the assistance of the fully labeled source domain, attracts increasing attention in recent years.
no code implementations • CVPR 2021 • Liyuan Wang, Kuo Yang, Chongxuan Li, Lanqing Hong, Zhenguo Li, Jun Zhu
Continual learning usually assumes the incoming data are fully labeled, which might not be applicable in real applications.
no code implementations • 1 Jan 2021 • Yimin Huang, YuJun Li, Zhenguo Li, Zhihua Zhang
Moreover, comparisons between different initial designs with the same model show the advantage of the proposed optimal design.
no code implementations • 1 Jan 2021 • Xu Lan, Steven McDonagh, Shaogang Gong, Jiali Wang, Zhenguo Li, Sarah Parisot
Semi-Supervised Few-shot Learning (SS-FSL) investigates the benefit of incorporating unlabelled data in few-shot settings.
no code implementations • 1 Jan 2021 • Yao Zhu, Jiacheng Sun, Zewei Chen, Zhenguo Li
We justify the algorithm with a linear model that the added saliency maps pull data away from its closest decision boundary.
2 code implementations • 1 Jan 2021 • Yawen Duan, Xin Chen, Hang Xu, Zewei Chen, Xiaodan Liang, Tong Zhang, Zhenguo Li
While existing NAS methods mostly design architectures on one single task, algorithms that look beyond single-task search are surging to pursue a more efficient and universal solution across various tasks.
no code implementations • 1 Jan 2021 • Kaichen Zhou, Lanqing Hong, Fengwei Zhou, Binxin Ru, Zhenguo Li, Trigoni Niki, Jiashi Feng
Our method performs co-optimization of the neural architectures, training hyper-parameters and data augmentation policies in an end-to-end fashion without the need of model retraining.
no code implementations • ICCV 2021 • Hanxue Liang, Chenhan Jiang, Dapeng Feng, Xin Chen, Hang Xu, Xiaodan Liang, Wei zhang, Zhenguo Li, Luc van Gool
Here we present a novel self-supervised 3D Object detection framework that seamlessly integrates the geometry-aware contrast and clustering harmonization to lift the unsupervised 3D representation learning, named GCC-3D.
no code implementations • 1 Jan 2021 • Hang Xu, Ning Kang, Gengwei Zhang, Xiaodan Liang, Zhenguo Li
The resulting model zoo is more training efficient than SOTA NAS models, e. g. 6x faster than RegNetY-16GF, and 1. 7x faster than EfficientNetB3.
no code implementations • 22 Dec 2020 • Fengwei Zhou, Jiawei Li, Chuanlong Xie, Fei Chen, Lanqing Hong, Rui Sun, Zhenguo Li
Automated data augmentation has shown superior performance in image recognition.
1 code implementation • 17 Dec 2020 • Haoyue Bai, Rui Sun, Lanqing Hong, Fengwei Zhou, Nanyang Ye, Han-Jia Ye, S. -H. Gary Chan, Zhenguo Li
To address that, we propose DecAug, a novel decomposed feature representation and semantic augmentation approach for OoD generalization.
1 code implementation • 16 Dec 2020 • Huifeng Guo, Bo Chen, Ruiming Tang, Weinan Zhang, Zhenguo Li, Xiuqiang He
In this paper, we propose a novel embedding learning framework for numerical features in CTR prediction (AutoDis) with high model capacity, end-to-end training and unique representation properties preserved.
no code implementations • 7 Dec 2020 • Gengwei Zhang, Yiming Gao, Hang Xu, Hao Zhang, Zhenguo Li, Xiaodan Liang
Panoptic segmentation that unifies instance segmentation and semantic segmentation has recently attracted increasing attention.
Ranked #14 on
Panoptic Segmentation
on COCO test-dev
no code implementations • 4 Dec 2020 • Xiao-Yun Zhou, Jiacheng Sun, Nanyang Ye, Xu Lan, Qijun Luo, Bo-Lin Lai, Pedro Esperanca, Guang-Zhong Yang, Zhenguo Li
Among previous normalization methods, Batch Normalization (BN) performs well at medium and large batch sizes and is with good generalizability to multiple vision tasks, while its performance degrades significantly at small batch sizes.
no code implementations • 18 Nov 2020 • Bo Xiong, Yimin Huang, Hanrong Ye, Steffen Staab, Zhenguo Li
MOFA pursues several rounds of HPO, where each round alternates between exploration of hyperparameter space by factorial design and exploitation of evaluation results by factorial analysis.
1 code implementation • 3 Nov 2020 • Bochao Wang, Hang Xu, Jiajin Zhang, Chen Chen, Xiaozhi Fang, Yixing Xu, Ning Kang, Lanqing Hong, Chenhan Jiang, Xinyue Cai, Jiawei Li, Fengwei Zhou, Yong Li, Zhicheng Liu, Xinghao Chen, Kai Han, Han Shu, Dehua Song, Yunhe Wang, Wei zhang, Chunjing Xu, Zhenguo Li, Wenzhi Liu, Tong Zhang
Automated Machine Learning (AutoML) is an important industrial solution for automatic discovery and deployment of the machine learning models.
no code implementations • 3 Sep 2020 • Qi Sun, Hexin Dong, Zewei Chen, Weizhen Dian, Jiacheng Sun, Yitong Sun, Zhenguo Li, Bin Dong
Gradient-based algorithms for training ResNets typically require a forward pass of the input data, followed by back-propagating the objective gradient to update parameters, which are time-consuming for deep ResNets.
1 code implementation • ECCV 2020 • Hang Xu, Shaoju Wang, Xinyue Cai, Wei zhang, Xiaodan Liang, Zhenguo Li
In this paper, we propose a novel lane-sensitive architecture search framework named CurveLane-NAS to automatically capture both long-ranged coherent and accurate short-range curve information while unifying both architecture search and post-processing on curve lane predictions via point blending.
Ranked #4 on
Lane Detection
on CurveLanes
no code implementations • ECCV 2020 • Wenshuo Ma, Tingzhong Tian, Hang Xu, Yimin Huang, Zhenguo Li
By carefully analyzing the existing bounding box patterns on the feature hierarchy, we design a flexible and tight hyper-parameter space for anchor configurations.
no code implementations • ECCV 2020 • Xin Chen, Yawen Duan, Zewei Chen, Hang Xu, Zihao Chen, Xiaodan Liang, Tong Zhang, Zhenguo Li
In spite of its remarkable progress, many algorithms are restricted to particular search spaces.
Ranked #4 on
Neural Architecture Search
on NAS-Bench-201, ImageNet-16-120
(Accuracy (val) metric)
1 code implementation • 11 Jul 2020 • Yimin Huang, Yu-Jun Li, Hanrong Ye, Zhenguo Li, Zhihua Zhang
The evaluation of hyperparameters, neural architectures, or data augmentation policies becomes a critical model selection problem in advanced deep learning with a large hyperparameter search space.
no code implementations • 2 Jul 2020 • Yifei Wang, Dan Peng, Furui Liu, Zhenguo Li, Zhitang Chen, Jiansheng Yang
Adversarial Training (AT) is proposed to alleviate the adversarial vulnerability of machine learning models by extracting only robust features from the input, which, however, inevitably leads to severe accuracy reduction as it discards the non-robust yet useful features.
no code implementations • 16 Jun 2020 • Jiacheng Sun, Xiangyong Cao, Hanwen Liang, Weiran Huang, Zewei Chen, Zhenguo Li
In recent years, a variety of normalization methods have been proposed to help train neural networks, such as batch normalization (BN), layer normalization (LN), weight normalization (WN), group normalization (GN), etc.
no code implementations • 13 Jun 2020 • Chuanlong Xie, Haotian Ye, Fei Chen, Yue Liu, Rui Sun, Zhenguo Li
The key of the out-of-distribution (OOD) generalization is to generalize invariance from training domains to target domains.
3 code implementations • NeurIPS 2020 • Kai Zheng, Tianle Cai, Weiran Huang, Zhenguo Li, Li-Wei Wang
We study locally differentially private (LDP) bandits learning in this paper.
no code implementations • CVPR 2020 • Aoxue Li, Weiran Huang, Xu Lan, Jiashi Feng, Zhenguo Li, Li-Wei Wang
Few-shot learning (FSL) has attracted increasing attention in recent years but remains challenging, due to the intrinsic difficulty in learning to generalize from a few examples.
Ranked #1 on
Few-Shot Image Classification
on ImageNet (1-shot)
1 code implementation • CVPR 2020 • Xiawu Zheng, Rongrong Ji, Qiang Wang, Qixiang Ye, Zhenguo Li, Yonghong Tian, Qi Tian
In this paper, we provide a novel yet systematic rethinking of PE in a resource constrained regime, termed budgeted PE (BPE), which precisely and effectively estimates the performance of an architecture sampled from an architecture space.
4 code implementations • 25 Mar 2020 • Bin Liu, Chenxu Zhu, Guilin Li, Wei-Nan Zhang, Jincai Lai, Ruiming Tang, Xiuqiang He, Zhenguo Li, Yong Yu
By implementing a regularized optimizer over the architecture parameters, the model can automatically identify and remove the redundant feature interactions during the training process of the model.
Ranked #15 on
Click-Through Rate Prediction
on Criteo
no code implementations • 18 Feb 2020 • Linpu Fang, Hang Xu, Zhili Liu, Sarah Parisot, Zhenguo Li
In this paper, we study the hybrid-supervised object detection problem, aiming to train a high quality detector with only a limited amount of fullyannotated data and fully exploiting cheap data with imagelevel labels.
no code implementations • 18 Feb 2020 • Hang Xu, Linpu Fang, Xiaodan Liang, Wenxiong Kang, Zhenguo Li
Finally, an InterDomain Transfer Module is proposed to exploit diverse transfer dependencies across all domains and enhance the regional feature representation by attending and transferring semantic contexts globally.
no code implementations • 23 Jan 2020 • Zewei Chen, Fengwei Zhou, George Trimponias, Zhenguo Li
Despite recent progress, the problem of approximating the full Pareto front accurately and efficiently remains challenging.
no code implementations • 22 Jan 2020 • Mi Luo, Fei Chen, Pengxiang Cheng, Zhenhua Dong, Xiuqiang He, Jiashi Feng, Zhenguo Li
Recommender systems often face heterogeneous datasets containing highly personalized historical data of users, where no single model could give the best recommendation for every user.
no code implementations • 24 Dec 2019 • Yimin Huang, Weiran Huang, Liang Li, Zhenguo Li
In this paper, we mainly consider the scenario in which we have a common model set used for model averaging instead of selecting a single final model via a model selection procedure to account for this model's uncertainty to improve reliability and accuracy of inferences.
no code implementations • 22 Nov 2019 • Lewei Yao, Hang Xu, Wei zhang, Xiaodan Liang, Zhenguo Li
In this paper, we present a two-stage coarse-to-fine searching strategy named Structural-to-Modular NAS (SM-NAS) for searching a GPU-friendly design of both an efficient combination of modules and better modular-level architecture for object detection.
1 code implementation • NeurIPS 2020 • Han Shi, Renjie Pi, Hang Xu, Zhenguo Li, James T. Kwok, Tong Zhang
In this work, we propose BONAS (Bayesian Optimized Neural Architecture Search), a sample-based NAS framework which is accelerated using weight-sharing to evaluate multiple related architectures simultaneously.
1 code implementation • 26 Sep 2019 • Guilin Li, Xing Zhang, Zitong Wang, Matthias Tan, Jiashi Feng, Zhenguo Li, Tong Zhang
Recently, the efficiency of automatic neural architecture design has been significantly improved by gradient-based search methods such as DARTS.
no code implementations • 25 Sep 2019 • Han Shi, Renjie Pi, Hang Xu, Zhenguo Li, James T. Kwok, Tong Zhang
Inspired by the nature of the graph structure of a neural network, we propose BOGCN-NAS, a NAS algorithm using Bayesian Optimization with Graph Convolutional Network (GCN) predictor.
no code implementations • 13 Sep 2019 • Hanwen Liang, Shifeng Zhang, Jiacheng Sun, Xingqiu He, Weiran Huang, Kechen Zhuang, Zhenguo Li
Therefore, we propose a simple and effective algorithm, named "DARTS+", to avoid the collapse and improve the original DARTS, by "early stopping" the search procedure when meeting a certain criterion.
1 code implementation • 16 May 2019 • Lin Lan, Zhenguo Li, Xiaohong Guan, Pinghui Wang
Despite significant progress, deep reinforcement learning (RL) suffers from data-inefficiency and limited generalization.
no code implementations • 28 Nov 2018 • Steven McDonagh, Sarah Parisot, Fengwei Zhou, Xing Zhang, Ales Leonardis, Zhenguo Li, Gregory Slabaugh
In this work, we propose a new approach that affords fast adaptation to previously unseen cameras, and robustness to changes in capture device by leveraging annotated samples across different cameras and datasets.
5 code implementations • 12 Apr 2018 • Huifeng Guo, Ruiming Tang, Yunming Ye, Zhenguo Li, Xiuqiang He, Zhenhua Dong
In this paper, we study two instances of DeepFM where its "deep" component is DNN and PNN respectively, for which we denote as DeepFM-D and DeepFM-P. Comprehensive experiments are conducted to demonstrate the effectiveness of DeepFM-D and DeepFM-P over the existing models for CTR prediction, on both benchmark data and commercial data.
no code implementations • 22 Feb 2018 • Fei Chen, Mi Luo, Zhenhua Dong, Zhenguo Li, Xiuqiang He
Statistical and systematic challenges in collaboratively training machine learning models across distributed networks of mobile devices have been the bottlenecks in the real-world application of federated learning.
no code implementations • 10 Feb 2018 • Fengwei Zhou, Bin Wu, Zhenguo Li
Few-shot learning remains challenging for meta-learning that learns a learning algorithm (meta-learner) from many related tasks.
1 code implementation • 13 Aug 2017 • Chenzi Zhang, Fan Wei, Qin Liu, Zhihao Gavin Tang, Zhenguo Li
We provide a worst-case upper bound of replication factor for our heuristic on general graphs.
7 code implementations • 31 Jul 2017 • Zhenguo Li, Fengwei Zhou, Fei Chen, Hang Li
In contrast, meta-learning learns from many related tasks a meta-learner that can learn a new task more accurately and faster with fewer examples, where the choice of meta-learners is crucial.
16 code implementations • 13 Mar 2017 • Huifeng Guo, Ruiming Tang, Yunming Ye, Zhenguo Li, Xiuqiang He
Learning sophisticated feature interactions behind user behaviors is critical in maximizing CTR for recommender systems.
Ranked #1 on
Click-Through Rate Prediction
on Company*
no code implementations • CVPR 2015 • Xiao-Ming Wu, Zhenguo Li, Shih-Fu Chang
Graph-based computer vision applications rely critically on similarity metrics which compute the pairwise similarity between any pair of vertices on graphs.
no code implementations • CVPR 2014 • Go Irie, Zhenguo Li, Xiao-Ming Wu, Shih-Fu Chang
Previous efforts in hashing intend to preserve data variance or pairwise affinity, but neither is adequate in capturing the manifold structures hidden in most visual data.
no code implementations • NeurIPS 2013 • Xiao-Ming Wu, Zhenguo Li, Shih-Fu Chang
We show that either explicitly or implicitly, various well-known graph-based models exhibit a common significant \emph{harmonic} structure in its target function -- the value of a vertex is approximately the weighted average of the values of its adjacent neighbors.
no code implementations • CVPR 2013 • Go Irie, Dong Liu, Zhenguo Li, Shih-Fu Chang
nary learning methods rely on image descriptors alone or together with class labels.
no code implementations • NeurIPS 2012 • Xiao-Ming Wu, Zhenguo Li, Anthony M. So, John Wright, Shih-Fu Chang
We prove that under proper absorption rates, a random walk starting from a set $\mathcal{S}$ of low conductance will be mostly absorbed in $\mathcal{S}$.
no code implementations • NeurIPS 2009 • Xiao-Ming Wu, Anthony M. So, Zhenguo Li, Shuo-Yen R. Li
In this paper, we show that a large class of kernel learning problems can be reformulated as semidefinite-quadratic-linear programs (SQLPs), which only contain a simple positive semidefinite constraint, a second-order cone constraint and a number of linear constraints.