arXiv subjects

Nojun Kwak

Publications and source records attributed to Nojun Kwak.

At least 145 records · Page 8Linked to original sources

arXiv2020

StackNet: Stacking Parameters for Continual learning

Training a neural network for a classification task typically assumes that the data to train are given from the beginning. However, in the real world, additional data accumulate gradually and the model requires additional training without accessing the old training data. This usually leads to the catastrophic forgetting problem which is inevitable for the traditional training methodology of neural networks. In this paper, we propose a continual learning method that is able to learn additional tasks while retaining the performance of previously learned tasks by stacking parameters. Composed of two complementary components, the index module and the StackNet, our method estimates the index of the corresponding task for an input sample with the index module and utilizes a particular portion of StackNet with this index. The StackNet guarantees no degradation in the performance of the previously learned tasks and the index module shows high confidence in finding the origin of an input sample. Compared to the previous work of PackNet, our method is competitive and highly intuitive.

cs.LG↗

arXiv2020

LSQ+: Improving low-bit quantization through learnable offsets and better initialization

Unlike ReLU, newer activation functions (like Swish, H-swish, Mish) that are frequently employed in popular efficient architectures can also result in negative activation values, with skewed positive and negative ranges. Typical learnable quantization schemes [PACT, LSQ] assume unsigned quantization for activations and quantize all negative activations to zero which leads to significant loss in performance. Naively using signed quantization to accommodate these negative values requires an extra sign bit which is expensive for low-bit (2-, 3-, 4-bit) quantization. To solve this problem, we propose LSQ+, a natural extension of LSQ, wherein we introduce a general asymmetric quantization scheme with trainable scale and offset parameters that can learn to accommodate the negative activations. Gradient-based learnable quantization schemes also commonly suffer from high instability or variance in the final training performance, hence requiring a great deal of hyper-parameter tuning to reach a satisfactory performance. LSQ+ alleviates this problem by using an MSE-based initialization scheme for the quantization parameters. We show that this initialization leads to significantly lower variance in final performance across multiple training runs. Overall, LSQ+ shows state-of-the-art results for EfficientNet and MixNet and also significantly outperforms LSQ for low-bit quantization of neural nets with Swish activations (e.g.: 1.8% gain with W4A4 quantization and upto 5.6% gain with W2A2 quantization of EfficientNet-B0 on ImageNet dataset). To the best of our knowledge, ours is the first work to quantize such architectures to extremely low bit-widths.

cs.CV↗

arXiv2020

Class-Imbalanced Semi-Supervised Learning

Semi-Supervised Learning (SSL) has achieved great success in overcoming the difficulties of labeling and making full use of unlabeled data. However, SSL has a limited assumption that the numbers of samples in different classes are balanced, and many SSL algorithms show lower performance for the datasets with the imbalanced class distribution. In this paper, we introduce a task of class-imbalanced semi-supervised learning (CISSL), which refers to semi-supervised learning with class-imbalanced data. In doing so, we consider class imbalance in both labeled and unlabeled sets. First, we analyze existing SSL methods in imbalanced environments and examine how the class imbalance affects SSL methods. Then we propose Suppressed Consistency Loss (SCL), a regularization method robust to class imbalance. Our method shows better performance than the conventional methods in the CISSL environment. In particular, the more severe the class imbalance and the smaller the size of the labeled data, the better our method performs.

cs.LG↗

arXiv2020

SINet: Extreme Lightweight Portrait Segmentation Networks with Spatial Squeeze Modules and Information Blocking Decoder

Designing a lightweight and robust portrait segmentation algorithm is an important task for a wide range of face applications. However, the problem has been considered as a subset of the object segmentation problem and less handled in the semantic segmentation field. Obviously, portrait segmentation has its unique requirements. First, because the portrait segmentation is performed in the middle of a whole process of many real-world applications, it requires extremely lightweight models. Second, there has not been any public datasets in this domain that contain a sufficient number of images with unbiased statistics. To solve the first problem, we introduce the new extremely lightweight portrait segmentation model SINet, containing an information blocking decoder and spatial squeeze modules. The information blocking decoder uses confidence estimates to recover local spatial information without spoiling global consistency. The spatial squeeze module uses multiple receptive fields to cope with various sizes of consistency in the image. To tackle the second problem, we propose a simple method to create additional portrait segmentation data which can improve accuracy on the EG1800 dataset. In our qualitative and quantitative analysis on the EG1800 dataset, we show that our method outperforms various existing lightweight segmentation models. Our method reduces the number of parameters from 2.1M to 86.9K (around 95.9% reduction), while maintaining the accuracy under an 1% margin from the state-of-the-art portrait segmentation method. We also show our model is successfully executed on a real mobile device with 100.6 FPS. In addition, we demonstrate that our method can be used for general semantic segmentation on the Cityscapes dataset. The code and dataset are available in https://github.com/HYOJINPARK/ExtPortraitSeg .

cs.CV↗

arXiv2019

ExtremeC3Net: Extreme Lightweight Portrait Segmentation Networks using Advanced C3-modules

Designing a lightweight and robust portrait segmentation algorithm is an important task for a wide range of face applications. However, the problem has been considered as a subset of the object segmentation problem. bviously, portrait segmentation has its unique requirements. First, because the portrait segmentation is performed in the middle of a whole process of many realworld applications, it requires extremely lightweight models. Second, there has not been any public datasets in this domain that contain a sufficient number of images with unbiased statistics. To solve the problems, we introduce a new extremely lightweight portrait segmentation model consisting of a two-branched architecture based on the concentrated-comprehensive convolutions block. Our method reduces the number of parameters from 2.1M to 37.7K (around 98.2% reduction), while maintaining the accuracy within a 1% margin from the state-of-the-art portrait segmentation method. In our qualitative and quantitative analysis on the EG1800 dataset, we show that our method outperforms various existing lightweight segmentation models. Second, we propose a simple method to create additional portrait segmentation data which can improve accuracy on the EG1800 dataset. Also, we analyze the bias in public datasets by additionally annotating race, gender, and age on our own. The augmented dataset, the additional annotations and code are available in https://github.com/HYOJINPARK/ExtPortraitSeg .

cs.CV↗

arXiv2019

QKD: Quantization-aware Knowledge Distillation

Quantization and Knowledge distillation (KD) methods are widely used to reduce memory and power consumption of deep neural networks (DNNs), especially for resource-constrained edge devices. Although their combination is quite promising to meet these requirements, it may not work as desired. It is mainly because the regularization effect of KD further diminishes the already reduced representation power of a quantized model. To address this short-coming, we propose Quantization-aware Knowledge Distillation (QKD) wherein quantization and KD are care-fully coordinated in three phases. First, Self-studying (SS) phase fine-tunes a quantized low-precision student network without KD to obtain a good initialization. Second, Co-studying (CS) phase tries to train a teacher to make it more quantizaion-friendly and powerful than a fixed teacher. Finally, Tutoring (TU) phase transfers knowledge from the trained teacher to the student. We extensively evaluate our method on ImageNet and CIFAR-10/100 datasets and show an ablation study on networks with both standard and depthwise-separable convolutions. The proposed QKD outperformed existing state-of-the-art methods (e.g., 1.3% improvement on ResNet-18 with W4A4, 2.6% on MobileNetV2 with W4A4). Additionally, QKD could recover the full-precision accuracy at as low as W3A3 quantization on ResNet and W6A6 quantization on MobilenetV2.

cs.CV↗

arXiv2019

Tell Me What They're Holding: Weakly-supervised Object Detection with Transferable Knowledge from Human-object Interaction

In this work, we introduce a novel weakly supervised object detection (WSOD) paradigm to detect objects belonging to rare classes that have not many examples using transferable knowledge from human-object interactions (HOI). While WSOD shows lower performance than full supervision, we mainly focus on HOI as the main context which can strongly supervise complex semantics in images. Therefore, we propose a novel module called RRPN (relational region proposal network) which outputs an object-localizing attention map only with human poses and action verbs. In the source domain, we fully train an object detector and the RRPN with full supervision of HOI. With transferred knowledge about localization map from the trained RRPN, a new object detector can learn unseen objects with weak verbal supervision of HOI without bounding box annotations in the target domain. Because the RRPN is designed as an add-on type, we can apply it not only to the object detection but also to other domains such as semantic segmentation. The experimental results on HICO-DET dataset show the possibility that the proposed method can be a cheap alternative for the current supervised object detection paradigm. Moreover, qualitative results demonstrate that our model can properly localize unseen objects on HICO-DET and V-COCO datasets.

cs.CV↗

arXiv2019

Sym-parameterized Dynamic Inference for Mixed-Domain Image Translation

Recent advances in image-to-image translation have led to some ways to generate multiple domain images through a single network. However, there is still a limit in creating an image of a target domain without a dataset on it. We propose a method that expands the concept of `multi-domain' from data to the loss area and learns the combined characteristics of each domain to dynamically infer translations of images in mixed domains. First, we introduce Sym-parameter and its learning method for variously mixed losses while synchronizing them with input conditions. Then, we propose Sym-parameterized Generative Network (SGN) which is empirically confirmed of learning mixed characteristics of various data and losses, and translating images to any mixed-domain without ground truths, such as 30% Van Gogh and 20% Monet and 40% snowy.

cs.CV↗

arXiv2019

FEED: Feature-level Ensemble for Knowledge Distillation

Knowledge Distillation (KD) aims to transfer knowledge in a teacher-student framework, by providing the predictions of the teacher network to the student network in the training stage to help the student network generalize better. It can use either a teacher with high capacity or {an} ensemble of multiple teachers. However, the latter is not convenient when one wants to use feature-map-based distillation methods. For a solution, this paper proposes a versatile and powerful training algorithm named FEature-level Ensemble for knowledge Distillation (FEED), which aims to transfer the ensemble knowledge using multiple teacher networks. We introduce a couple of training algorithms that transfer ensemble knowledge to the student at the feature map level. Among the feature-map-based distillation methods, using several non-linear transformations in parallel for transferring the knowledge of the multiple teacher{s} helps the student find more generalized solutions. We name this method as parallel FEED, andexperimental results on CIFAR-100 and ImageNet show that our method has clear performance enhancements, without introducing any additional parameters or computations at test time. We also show the experimental results of sequentially feeding teacher's information to the student, hence the name sequential FEED, and discuss the lessons obtained. Additionally, the empirical results on measuring the reconstruction errors at the feature map give hints for the enhancements.

cs.CV↗

arXiv2019

A Comprehensive Overhaul of Feature Distillation

We investigate the design aspects of feature distillation methods achieving network compression and propose a novel feature distillation method in which the distillation loss is designed to make a synergy among various aspects: teacher transform, student transform, distillation feature position and distance function. Our proposed distillation loss includes a feature transform with a newly designed margin ReLU, a new distillation feature position, and a partial L2 distance function to skip redundant information giving adverse effects to the compression of student. In ImageNet, our proposed method achieves 21.65% of top-1 error with ResNet50, which outperforms the performance of the teacher network, ResNet152. Our proposed method is evaluated on various tasks such as image classification, object detection and semantic segmentation and achieves a significant performance improvement in all tasks. The code is available at https://sites.google.com/view/byeongho-heo/overhaul

cs.CV↗

arXiv2019

C3: Concentrated-Comprehensive Convolution and its application to semantic segmentation

One of the practical choices for making a lightweight semantic segmentation model is to combine a depth-wise separable convolution with a dilated convolution. However, the simple combination of these two methods results in an over-simplified operation which causes severe performance degradation due to loss of information contained in the feature map. To resolve this problem, we propose a new block called Concentrated-Comprehensive Convolution (C3) which applies the asymmetric convolutions before the depth-wise separable dilated convolution to compensate for the information loss due to dilated convolution. The C3 block consists of a concentration stage and a comprehensive convolution stage. The first stage uses two depth-wise asymmetric convolutions for compressed information from the neighboring pixels to alleviate the information loss. The second stage increases the receptive field by using a depth-wise separable dilated convolution from the feature map of the first stage. We applied the C3 block to various segmentation frameworks (ESPNet, DRN, ERFNet, ENet) for proving the beneficial properties of our proposed method. Experimental results show that the proposed method preserves the original accuracies on Cityscapes dataset while reducing the complexity. Furthermore, we modified ESPNet to achieve about 2% better performance while reducing the number of parameters by half and the number of FLOPs by 35% compared with the original ESPNet. Finally, experiments on ImageNet classification task show that C3 block can successfully replace dilated convolutions.

cs.CV↗

arXiv2019

Image Enhancement by Recurrently-trained Super-resolution Network

We introduce a new learning strategy for image enhancement by recurrently training the same simple superresolution (SR) network multiple times. After initially training an SR network by using pairs of a corrupted low resolution (LR) image and an original image, the proposed method makes use of the trained SR network to generate new high resolution (HR) images with a doubled resolution from the original uncorrupted images. Then, the new HR images are downscaled to the original resolution, which work as target images for the SR network in the next stage. The newly generated HR images by the repeatedly trained SR network show better image quality and this strategy of training LR to mimic new HR can lead to a more efficient SR network. Up to a certain point, by repeating this process multiple times, better and better images are obtained. This recurrent leaning strategy for SR can be a good solution for downsizing convolution networks and making a more efficient SR network. To measure the enhanced image quality, for the first time in this area of super-resolution and image enhancement, we use VIQET MOS score which reflects human visual quality more accurately than the conventional MSE measure.

eess.IV↗

arXiv2019

Textbook Question Answering with Multi-modal Context Graph Understanding and Self-supervised Open-set Comprehension

In this work, we introduce a novel algorithm for solving the textbook question answering (TQA) task which describes more realistic QA problems compared to other recent tasks. We mainly focus on two related issues with analysis of the TQA dataset. First, solving the TQA problems requires to comprehend multi-modal contexts in complicated input data. To tackle this issue of extracting knowledge features from long text lessons and merging them with visual features, we establish a context graph from texts and images, and propose a new module f-GCN based on graph convolutional networks (GCN). Second, scientific terms are not spread over the chapters and subjects are split in the TQA dataset. To overcome this so called "out-of-domain" issue, before learning QA problems, we introduce a novel self-supervised open-set learning process without any annotations. The experimental results show that our model significantly outperforms prior state-of-the-art methods. Moreover, ablation studies validate that both methods of incorporating f-GCN for extracting knowledge from multi-modal contexts and our newly proposed self-supervised learning process are effective for TQA problems.

cs.CL↗

arXiv2019

Pose estimator and tracker using temporal flow maps for limbs

For human pose estimation in videos, it is significant how to use temporal information between frames. In this paper, we propose temporal flow maps for limbs (TML) and a multi-stride method to estimate and track human poses. The proposed temporal flow maps are unit vectors describing the limbs' movements. We constructed a network to learn both spatial information and temporal information end-to-end. Spatial information such as joint heatmaps and part affinity fields is regressed in the spatial network part, and the TML is regressed in the temporal network part. We also propose a data augmentation method to learn various types of TML better. The proposed multi-stride method expands the data by randomly selecting two frames within a defined range. We demonstrate that the proposed method efficiently estimates and tracks human poses on the PoseTrack 2017 and 2018 datasets.

cs.CV↗

arXiv2019

Disentangling Options with Hellinger Distance Regularizer

In reinforcement learning (RL), temporal abstraction still remains as an important and unsolved problem. The options framework provided clues to temporal abstraction in the RL, and the option-critic architecture elegantly solved the two problems of finding options and learning RL agents in an end-to-end manner. However, it is necessary to examine whether the options learned through this method play a mutually exclusive role. In this paper, we propose a Hellinger distance regularizer, a method for disentangling options. In addition, we will shed light on various indicators from the statistical point of view to compare with the options learned through the existing option-critic architecture.

cs.LG↗

arXiv2019

URNet : User-Resizable Residual Networks with Conditional Gating Module

Convolutional Neural Networks are widely used to process spatial scenes, but their computational cost is fixed and depends on the structure of the network used. There are methods to reduce the cost by compressing networks or varying its computational path dynamically according to the input image. However, since a user can not control the size of the learned model, it is difficult to respond dynamically if the amount of service requests suddenly increases. We propose User-Resizable Residual Networks (URNet), which allows users to adjust the scale of the network as needed during evaluation. URNet includes Conditional Gating Module (CGM) that determines the use of each residual block according to the input image and the desired scale. CGM is trained in a supervised manner using the newly proposed scale loss and its corresponding training methods. URNet can control the amount of computation according to user's demand without degrading the accuracy significantly. It can also be used as a general compression method by fixing the scale size during training. In the experiments on ImageNet, URNet based on ResNet-101 maintains the accuracy of the baseline even when resizing it to approximately 80% of the original network, and demonstrates only about 1% accuracy degradation when using about 65% of the computation.

cs.CV↗

arXiv2019

Task-oriented Design through Deep Reinforcement Learning

We propose a new low-cost machine-learning-based methodology which assists designers in reducing the gap between the problem and the solution in the design process. Our work applies reinforcement learning (RL) to find the optimal task-oriented design solution through the construction of the design action for each task. For this task-oriented design, the 3D design process in product design is assigned to an action space in Deep RL, and the desired 3D model is obtained by training each design action according to the task. By showing that this method achieves satisfactory design even when applied to a task pursuing multiple goals, we suggest the direction of how machine learning can contribute to the design process. Also, we have validated with product designers that this methodology can assist the creative part in the process of design.

cs.LG↗

arXiv2018

Genetic-Gated Networks for Deep Reinforcement

We introduce the Genetic-Gated Networks (G2Ns), simple neural networks that combine a gate vector composed of binary genetic genes in the hidden layer(s) of networks. Our method can take both advantages of gradient-free optimization and gradient-based optimization methods, of which the former is effective for problems with multiple local minima, while the latter can quickly find local minima. In addition, multiple chromosomes can define different models, making it easy to construct multiple models and can be effectively applied to problems that require multiple models. We show that this G2N can be applied to typical reinforcement learning algorithms to achieve a large improvement in sample efficiency and performance.

cs.NE↗