Paper Group AWR 230
Placental Flattening via Volumetric Parameterization. Hate Speech Detection on Vietnamese Social Media Text using the Bidirectional-LSTM Model. REPAIR: Removing Representation Bias by Dataset Resampling. SUMBT: Slot-Utterance Matching for Universal and Scalable Belief Tracking. DISN: Deep Implicit Surface Network for High-quality Single-view 3D Rec …
Placental Flattening via Volumetric Parameterization
| Title | Placental Flattening via Volumetric Parameterization |
| Authors | S. Mazdak Abulnaga, Esra Abaci Turk, Mikhail Bessmeltsev, P. Ellen Grant, Justin Solomon, Polina Golland |
| Abstract | We present a volumetric mesh-based algorithm for flattening the placenta to a canonical template to enable effective visualization of local anatomy and function. Monitoring placental function in vivo promises to support pregnancy assessment and to improve care outcomes. We aim to alleviate visualization and interpretation challenges presented by the shape of the placenta when it is attached to the curved uterine wall. To do so, we flatten the volumetric mesh that captures placental shape to resemble the well-studied ex vivo shape. We formulate our method as a map from the in vivo shape to a flattened template that minimizes the symmetric Dirichlet energy to control distortion throughout the volume. Local injectivity is enforced via constrained line search during gradient descent. We evaluate the proposed method on 28 placenta shapes extracted from MRI images in a clinical study of placental function. We achieve sub-voxel accuracy in mapping the boundary of the placenta to the template while successfully controlling distortion throughout the volume. We illustrate how the resulting mapping of the placenta enhances visualization of placental anatomy and function. Our code is freely available at https://github.com/mabulnaga/placenta-flattening . |
| Tasks | |
| Published | 2019-03-12 |
| URL | https://arxiv.org/abs/1903.05044v3 |
| https://arxiv.org/pdf/1903.05044v3.pdf | |
| PWC | https://paperswithcode.com/paper/placental-flattening-via-volumetric |
| Repo | https://github.com/mabulnaga/placenta-flattening |
| Framework | none |
Hate Speech Detection on Vietnamese Social Media Text using the Bidirectional-LSTM Model
| Title | Hate Speech Detection on Vietnamese Social Media Text using the Bidirectional-LSTM Model |
| Authors | Hang Thi-Thuy Do, Huy Duc Huynh, Kiet Van Nguyen, Ngan Luu-Thuy Nguyen, Anh Gia-Tuan Nguyen |
| Abstract | In this paper, we describe our system which participates in the shared task of Hate Speech Detection on Social Networks of VLSP 2019 evaluation campaign. We are provided with the pre-labeled dataset and an unlabeled dataset for social media comments or posts. Our mission is to pre-process and build machine learning models to classify comments/posts. In this report, we use Bidirectional Long Short-Term Memory to build the model that can predict labels for social media text according to Clean, Offensive, Hate. With this system, we achieve comparative results with 71.43% on the public standard test set of VLSP 2019. |
| Tasks | Hate Speech Detection |
| Published | 2019-11-09 |
| URL | https://arxiv.org/abs/1911.03648v1 |
| https://arxiv.org/pdf/1911.03648v1.pdf | |
| PWC | https://paperswithcode.com/paper/hate-speech-detection-on-vietnamese-social |
| Repo | https://github.com/huynhduchuydp36/VLSP2019-SHARED-Task-Hate-Speech-Detection-on-Social-Networks-Using-Bi-Lstm |
| Framework | none |
REPAIR: Removing Representation Bias by Dataset Resampling
| Title | REPAIR: Removing Representation Bias by Dataset Resampling |
| Authors | Yi Li, Nuno Vasconcelos |
| Abstract | Modern machine learning datasets can have biases for certain representations that are leveraged by algorithms to achieve high performance without learning to solve the underlying task. This problem is referred to as “representation bias”. The question of how to reduce the representation biases of a dataset is investigated and a new dataset REPresentAtion bIas Removal (REPAIR) procedure is proposed. This formulates bias minimization as an optimization problem, seeking a weight distribution that penalizes examples easy for a classifier built on a given feature representation. Bias reduction is then equated to maximizing the ratio between the classification loss on the reweighted dataset and the uncertainty of the ground-truth class labels. This is a minimax problem that REPAIR solves by alternatingly updating classifier parameters and dataset resampling weights, using stochastic gradient descent. An experimental set-up is also introduced to measure the bias of any dataset for a given representation, and the impact of this bias on the performance of recognition models. Experiments with synthetic and action recognition data show that dataset REPAIR can significantly reduce representation bias, and lead to improved generalization of models trained on REPAIRed datasets. The tools used for characterizing representation bias, and the proposed dataset REPAIR algorithm, are available at https://github.com/JerryYLi/Dataset-REPAIR/. |
| Tasks | Temporal Action Localization |
| Published | 2019-04-16 |
| URL | http://arxiv.org/abs/1904.07911v1 |
| http://arxiv.org/pdf/1904.07911v1.pdf | |
| PWC | https://paperswithcode.com/paper/190407911 |
| Repo | https://github.com/JerryYLi/Dataset-REPAIR |
| Framework | pytorch |
SUMBT: Slot-Utterance Matching for Universal and Scalable Belief Tracking
| Title | SUMBT: Slot-Utterance Matching for Universal and Scalable Belief Tracking |
| Authors | Hwaran Lee, Jinsik Lee, Tae-Yoon Kim |
| Abstract | In goal-oriented dialog systems, belief trackers estimate the probability distribution of slot-values at every dialog turn. Previous neural approaches have modeled domain- and slot-dependent belief trackers, and have difficulty in adding new slot-values, resulting in lack of flexibility of domain ontology configurations. In this paper, we propose a new approach to universal and scalable belief tracker, called slot-utterance matching belief tracker (SUMBT). The model learns the relations between domain-slot-types and slot-values appearing in utterances through attention mechanisms based on contextual semantic vectors. Furthermore, the model predicts slot-value labels in a non-parametric way. From our experiments on two dialog corpora, WOZ 2.0 and MultiWOZ, the proposed model showed performance improvement in comparison with slot-dependent methods and achieved the state-of-the-art joint accuracy. |
| Tasks | Goal-Oriented Dialog |
| Published | 2019-07-17 |
| URL | https://arxiv.org/abs/1907.07421v1 |
| https://arxiv.org/pdf/1907.07421v1.pdf | |
| PWC | https://paperswithcode.com/paper/sumbt-slot-utterance-matching-for-universal |
| Repo | https://github.com/SKTBrain/SUMBT |
| Framework | pytorch |
DISN: Deep Implicit Surface Network for High-quality Single-view 3D Reconstruction
| Title | DISN: Deep Implicit Surface Network for High-quality Single-view 3D Reconstruction |
| Authors | Qiangeng Xu, Weiyue Wang, Duygu Ceylan, Radomir Mech, Ulrich Neumann |
| Abstract | Reconstructing 3D shapes from single-view images has been a long-standing research problem. In this paper, we present DISN, a Deep Implicit Surface Network which can generate a high-quality detail-rich 3D mesh from an 2D image by predicting the underlying signed distance fields. In addition to utilizing global image features, DISN predicts the projected location for each 3D point on the 2D image, and extracts local features from the image feature maps. Combining global and local features significantly improves the accuracy of the signed distance field prediction, especially for the detail-rich areas. To the best of our knowledge, DISN is the first method that constantly captures details such as holes and thin structures present in 3D shapes from single-view images. DISN achieves the state-of-the-art single-view reconstruction performance on a variety of shape categories reconstructed from both synthetic and real images. Code is available at https://github.com/xharlie/DISN The supplementary can be found at https://xharlie.github.io/images/neurips_2019_supp.pdf |
| Tasks | 3D Reconstruction, Single-View 3D Reconstruction |
| Published | 2019-05-26 |
| URL | https://arxiv.org/abs/1905.10711v2 |
| https://arxiv.org/pdf/1905.10711v2.pdf | |
| PWC | https://paperswithcode.com/paper/disn-deep-implicit-surface-network-for-high |
| Repo | https://github.com/laughtervv/DISN |
| Framework | tf |
SDM: Sequential Deep Matching Model for Online Large-scale Recommender System
| Title | SDM: Sequential Deep Matching Model for Online Large-scale Recommender System |
| Authors | Fuyu Lv, Taiwei Jin, Changlong Yu, Fei Sun, Quan Lin, Keping Yang, Wilfred Ng |
| Abstract | Capturing users’ precise preferences is a fundamental problem in large-scale recommender system. Currently, item-based Collaborative Filtering (CF) methods are common matching approaches in industry. However, they are not effective to model dynamic and evolving preferences of users. In this paper, we propose a new sequential deep matching (SDM) model to capture users’ dynamic preferences by combining short-term sessions and long-term behaviors. Compared with existing sequence-aware recommendation methods, we tackle the following two inherent problems in real-world applications: (1) there could exist multiple interest tendencies in one session. (2) long-term preferences may not be effectively fused with current session interests. Long-term behaviors are various and complex, hence those highly related to the short-term session should be kept for fusion. We propose to encode behavior sequences with two corresponding components: multi-head self-attention module to capture multiple types of interests and long-short term gated fusion module to incorporate long-term preferences. Successive items are recommended after matching between sequential user behavior vector and item embedding vectors. Offline experiments on real-world datasets show the superior performance of the proposed SDM. Moreover, SDM has been successfully deployed on online large-scale recommender system at Taobao and achieves improvements in terms of a range of commercial metrics. |
| Tasks | Recommendation Systems |
| Published | 2019-09-01 |
| URL | https://arxiv.org/abs/1909.00385v2 |
| https://arxiv.org/pdf/1909.00385v2.pdf | |
| PWC | https://paperswithcode.com/paper/sdm-sequential-deep-matching-model-for-online |
| Repo | https://github.com/alicogintel/SDM |
| Framework | tf |
Representation Quality Explains Adversarial Attacks
| Title | Representation Quality Explains Adversarial Attacks |
| Authors | Danilo Vasconcellos Vargas, Shashank Kotyan, Moe Matsuki |
| Abstract | Neural networks have been shown vulnerable to adversarial samples, revealing that the representation learned is perhaps not as good as previously thought. Being able to evaluate to what extent are current neural networks’ representations capturing the existing features would be an essential step to understand the reason for the lack of robustness. Here, we propose a way to evaluate the representation quality of neural networks using a novel type of zero-shot test, entitled Raw Zero-Shot. The main idea lies in the fact that if an algorithm learns rich features, such features should be able to describe unknown classes as well. This happens because unknown classes usually share many common features with known classes if the features learned are general enough. Two metrics are proposed to evaluate the learned features. One is based on clustering validation techniques (Davies-Bouldin Index), and the other is based on the distance to an approximated ground-truth created automatically. Experiments reveal a strong correlation between such metrics and the robustness to adversarial attacks which is further supported by a high Pearson correlation and low p-value. Interestingly, the results suggest that dynamic routing networks such as CapsNet have better representation while current deeper DNNs are trading off representational quality for accuracy. Code available at http://bit.ly/RepresentationMetrics. |
| Tasks | |
| Published | 2019-06-15 |
| URL | https://arxiv.org/abs/1906.06627v4 |
| https://arxiv.org/pdf/1906.06627v4.pdf | |
| PWC | https://paperswithcode.com/paper/uncovering-why-deep-neural-networks-lack |
| Repo | https://github.com/shashankkotyan/RepresentationMetrics |
| Framework | tf |
Scaling Video Analytics on Constrained Edge Nodes
| Title | Scaling Video Analytics on Constrained Edge Nodes |
| Authors | Christopher Canel, Thomas Kim, Giulio Zhou, Conglong Li, Hyeontaek Lim, David G. Andersen, Michael Kaminsky, Subramanya R. Dulloor |
| Abstract | As video camera deployments continue to grow, the need to process large volumes of real-time data strains wide area network infrastructure. When per-camera bandwidth is limited, it is infeasible for applications such as traffic monitoring and pedestrian tracking to offload high-quality video streams to a datacenter. This paper presents FilterForward, a new edge-to-cloud system that enables datacenter-based applications to process content from thousands of cameras by installing lightweight edge filters that backhaul only relevant video frames. FilterForward introduces fast and expressive per-application microclassifiers that share computation to simultaneously detect dozens of events on computationally constrained edge nodes. Only matching events are transmitted to the cloud. Evaluation on two real-world camera feed datasets shows that FilterForward reduces bandwidth use by an order of magnitude while improving computational efficiency and event detection accuracy for challenging video content. |
| Tasks | |
| Published | 2019-05-24 |
| URL | https://arxiv.org/abs/1905.13536v1 |
| https://arxiv.org/pdf/1905.13536v1.pdf | |
| PWC | https://paperswithcode.com/paper/190513536 |
| Repo | https://github.com/viscloud/ff |
| Framework | none |
Personalised aesthetics with residual adapters
| Title | Personalised aesthetics with residual adapters |
| Authors | Carlos Rodríguez - Pardo, Hakan Bilen |
| Abstract | The use of computational methods to evaluate aesthetics in photography has gained interest in recent years due to the popularization of convolutional neural networks and the availability of new annotated datasets. Most studies in this area have focused on designing models that do not take into account individual preferences for the prediction of the aesthetic value of pictures. We propose a model based on residual learning that is capable of learning subjective, user specific preferences over aesthetics in photography, while surpassing the state-of-the-art methods and keeping a limited number of user-specific parameters in the model. Our model can also be used for picture enhancement, and it is suitable for content-based or hybrid recommender systems in which the amount of computational resources is limited. |
| Tasks | Recommendation Systems |
| Published | 2019-07-08 |
| URL | https://arxiv.org/abs/1907.03802v1 |
| https://arxiv.org/pdf/1907.03802v1.pdf | |
| PWC | https://paperswithcode.com/paper/personalised-aesthetics-with-residual-1 |
| Repo | https://github.com/crp94/Personalised-aesthetic-assessment-using-residual-adapters |
| Framework | pytorch |
Pontryagin Differentiable Programming: An End-to-End Learning and Control Framework
| Title | Pontryagin Differentiable Programming: An End-to-End Learning and Control Framework |
| Authors | Wanxin Jin, Zhaoran Wang, Zhuoran Yang, Shaoshuai Mou |
| Abstract | This paper develops a Pontryagin differentiable programming (PDP) methodology to establish a unified end-to-end learning framework, which solves a large class of learning and control tasks. The proposed PDP framework distinguishes itself from existing ones by two key techniques: first, by differentiating the Pontryagin’s Maximum Principle, the PDP framework allows end-to-end learning of a large class of parameterized systems, even when differentiation with respect to an unknown objective function is not readily attainable; and second, based on control theory, the PDP framework incorporates both the forward and backward propagations by constructing two control systems, respectively, which are then efficiently solved using techniques in control domain. Three learning modes of the proposed PDP framework are investigated to address three types of learning problems: inverse optimization, system identification, and control/planning, respectively. Effectiveness of the PDP framework in each learning mode has been validated in the context of pendulum systems. |
| Tasks | |
| Published | 2019-12-30 |
| URL | https://arxiv.org/abs/1912.12970v3 |
| https://arxiv.org/pdf/1912.12970v3.pdf | |
| PWC | https://paperswithcode.com/paper/pontryagin-differentiable-programming-an-end |
| Repo | https://github.com/wanxinjin/Pontryagin-Differentiable-Programming |
| Framework | none |
Emotion Detection in Text: Focusing on Latent Representation
| Title | Emotion Detection in Text: Focusing on Latent Representation |
| Authors | Armin Seyeditabari, Narges Tabari, Shafie Gholizadeh, Wlodek Zadrozny |
| Abstract | In recent years, emotion detection in text has become more popular due to its vast potential applications in marketing, political science, psychology, human-computer interaction, artificial intelligence, etc. In this work, we argue that current methods which are based on conventional machine learning models cannot grasp the intricacy of emotional language by ignoring the sequential nature of the text, and the context. These methods, therefore, are not sufficient to create an applicable and generalizable emotion detection methodology. Understanding these limitations, we present a new network based on a bidirectional GRU model to show that capturing more meaningful information from text can significantly improve the performance of these models. The results show significant improvement with an average of 26.8 point increase in F-measure on our test data and 38.6 increase on the totally new dataset. |
| Tasks | |
| Published | 2019-07-22 |
| URL | https://arxiv.org/abs/1907.09369v1 |
| https://arxiv.org/pdf/1907.09369v1.pdf | |
| PWC | https://paperswithcode.com/paper/emotion-detection-in-text-focusing-on-latent |
| Repo | https://github.com/armintabari/Emotion-Detection-RNN |
| Framework | tf |
A Robust Learning Approach to Domain Adaptive Object Detection
| Title | A Robust Learning Approach to Domain Adaptive Object Detection |
| Authors | Mehran Khodabandeh, Arash Vahdat, Mani Ranjbar, William G. Macready |
| Abstract | Domain shift is unavoidable in real-world applications of object detection. For example, in self-driving cars, the target domain consists of unconstrained road environments which cannot all possibly be observed in training data. Similarly, in surveillance applications sufficiently representative training data may be lacking due to privacy regulations. In this paper, we address the domain adaptation problem from the perspective of robust learning and show that the problem may be formulated as training with noisy labels. We propose a robust object detection framework that is resilient to noise in bounding box class labels, locations and size annotations. To adapt to the domain shift, the model is trained on the target domain using a set of noisy object bounding boxes that are obtained by a detection model trained only in the source domain. We evaluate the accuracy of our approach in various source/target domain pairs and demonstrate that the model significantly improves the state-of-the-art on multiple domain adaptation scenarios on the SIM10K, Cityscapes and KITTI datasets. |
| Tasks | Domain Adaptation, Object Detection, Robust Object Detection, Self-Driving Cars |
| Published | 2019-04-04 |
| URL | https://arxiv.org/abs/1904.02361v3 |
| https://arxiv.org/pdf/1904.02361v3.pdf | |
| PWC | https://paperswithcode.com/paper/a-robust-learning-approach-to-domain-adaptive |
| Repo | https://github.com/mkhodabandeh/robust_domain_adaptation |
| Framework | none |
Let’s FACE it. Finnish Poetry Generation with Aesthetics and Framing
| Title | Let’s FACE it. Finnish Poetry Generation with Aesthetics and Framing |
| Authors | Mika Hämäläinen, Khalid Alnajjar |
| Abstract | We present a creative poem generator for the morphologically rich Finnish language. Our method falls into the master-apprentice paradigm, where a computationally creative genetic algorithm teaches a BRNN model to generate poetry. We model several parts of poetic aesthetics in the fitness function of the genetic algorithm, such as sonic features, semantic coherence, imagery and metaphor. Furthermore, we justify the creativity of our method based on the FACE theory on computational creativity and take additional care in evaluating our system by automatic metrics for concepts together with human evaluation for aesthetics, framing and expressions. |
| Tasks | |
| Published | 2019-10-30 |
| URL | https://arxiv.org/abs/1910.13946v1 |
| https://arxiv.org/pdf/1910.13946v1.pdf | |
| PWC | https://paperswithcode.com/paper/lets-face-it-finnish-poetry-generation-with |
| Repo | https://github.com/mikahama/finmeter |
| Framework | tf |
Light-Weight RetinaNet for Object Detection
| Title | Light-Weight RetinaNet for Object Detection |
| Authors | Yixing Li, Fengbo Ren |
| Abstract | Object detection has gained great progress driven by the development of deep learning. Compared with a widely studied task – classification, generally speaking, object detection even need one or two orders of magnitude more FLOPs (floating point operations) in processing the inference task. To enable a practical application, it is essential to explore effective runtime and accuracy trade-off scheme. Recently, a growing number of studies are intended for object detection on resource constraint devices, such as YOLOv1, YOLOv2, SSD, MobileNetv2-SSDLite, whose accuracy on COCO test-dev detection results are yield to mAP around 22-25% (mAP-20-tier). On the contrary, very few studies discuss the computation and accuracy trade-off scheme for mAP-30-tier detection networks. In this paper, we illustrate the insights of why RetinaNet gives effective computation and accuracy trade-off for object detection and how to build a light-weight RetinaNet. We propose to only reduce FLOPs in computational intensive layers and keep other layer the same. Compared with most common way – input image scaling for FLOPs-accuracy trade-off, the proposed solution shows a constantly better FLOPs-mAP trade-off line. Quantitatively, the proposed method result in 0.1% mAP improvement at 1.15x FLOPs reduction and 0.3% mAP improvement at 1.8x FLOPs reduction. |
| Tasks | Object Detection |
| Published | 2019-05-24 |
| URL | https://arxiv.org/abs/1905.10011v1 |
| https://arxiv.org/pdf/1905.10011v1.pdf | |
| PWC | https://paperswithcode.com/paper/light-weight-retinanet-for-object-detection |
| Repo | https://github.com/PSCLab-ASU/LW-RetinaNet |
| Framework | none |
Deep Learning Theory Review: An Optimal Control and Dynamical Systems Perspective
| Title | Deep Learning Theory Review: An Optimal Control and Dynamical Systems Perspective |
| Authors | Guan-Horng Liu, Evangelos A. Theodorou |
| Abstract | Attempts from different disciplines to provide a fundamental understanding of deep learning have advanced rapidly in recent years, yet a unified framework remains relatively limited. In this article, we provide one possible way to align existing branches of deep learning theory through the lens of dynamical system and optimal control. By viewing deep neural networks as discrete-time nonlinear dynamical systems, we can analyze how information propagates through layers using mean field theory. When optimization algorithms are further recast as controllers, the ultimate goal of training processes can be formulated as an optimal control problem. In addition, we can reveal convergence and generalization properties by studying the stochastic dynamics of optimization algorithms. This viewpoint features a wide range of theoretical study from information bottleneck to statistical physics. It also provides a principled way for hyper-parameter tuning when optimal control theory is introduced. Our framework fits nicely with supervised learning and can be extended to other learning problems, such as Bayesian learning, adversarial training, and specific forms of meta learning, without efforts. The review aims to shed lights on the importance of dynamics and optimal control when developing deep learning theory. |
| Tasks | Meta-Learning |
| Published | 2019-08-28 |
| URL | https://arxiv.org/abs/1908.10920v2 |
| https://arxiv.org/pdf/1908.10920v2.pdf | |
| PWC | https://paperswithcode.com/paper/deep-learning-theory-review-an-optimal |
| Repo | https://github.com/ghliu/mean-field-fcdnn |
| Framework | pytorch |
