Paper Group ANR 1400
Fitness Dependent Optimizer: Inspired by the Bee Swarming Reproductive Process. Determining Relative Argument Specificity and Stance for Complex Argumentative Structures. CLEVR-Ref+: Diagnosing Visual Reasoning with Referring Expressions. OCGAN: One-class Novelty Detection Using GANs with Constrained Latent Representations. Identification of gateke …
Fitness Dependent Optimizer: Inspired by the Bee Swarming Reproductive Process
| Title | Fitness Dependent Optimizer: Inspired by the Bee Swarming Reproductive Process |
| Authors | Jaza M. Abdullah, Tarik A. Rashid |
| Abstract | In this paper, a novel swarm intelligent algorithm is proposed, known as the fitness dependent optimizer (FDO). The bee swarming reproductive process and their collective decision-making have inspired this algorithm; it has no algorithmic connection with the honey bee algorithm or the artificial bee colony algorithm. It is worth mentioning that FDO is considered a particle swarm optimization (PSO)-based algorithm that updates the search agent position by adding velocity (pace). However, FDO calculates velocity differently; it uses the problem fitness function value to produce weights, and these weights guide the search agents during both the exploration and exploitation phases. Throughout the paper, the FDO algorithm is presented, and the motivation behind the idea is explained. Moreover, FDO is tested on a group of 19 classical benchmark test functions, and the results are compared with three well-known algorithms: PSO, the genetic algorithm (GA), and the dragonfly algorithm (DA), additionally, FDO is tested on IEEE Congress of Evolutionary Computation Benchmark Test Functions (CEC-C06, 2019 Competition) [1]. The results are compared with three modern algorithms: (DA), the whale optimization algorithm (WOA), and the salp swarm algorithm (SSA). The FDO results show better performance in most cases and comparative results in other cases. Furthermore, the results are statistically tested with the Wilcoxon rank-sum test to show the significance of the results. Likewise, FDO stability in both the exploration and exploitation phases is verified and performance-proofed using different standard measurements. Finally, FDO is applied to real-world applications as evidence of its feasibility. |
| Tasks | Decision Making |
| Published | 2019-04-10 |
| URL | http://arxiv.org/abs/1904.05226v1 |
| http://arxiv.org/pdf/1904.05226v1.pdf | |
| PWC | https://paperswithcode.com/paper/fitness-dependent-optimizer-inspired-by-the |
| Repo | |
| Framework | |
Determining Relative Argument Specificity and Stance for Complex Argumentative Structures
| Title | Determining Relative Argument Specificity and Stance for Complex Argumentative Structures |
| Authors | Esin Durmus, Faisal Ladhak, Claire Cardie |
| Abstract | Systems for automatic argument generation and debate require the ability to (1) determine the stance of any claims employed in the argument and (2) assess the specificity of each claim relative to the argument context. Existing work on understanding claim specificity and stance, however, has been limited to the study of argumentative structures that are relatively shallow, most often consisting of a single claim that directly supports or opposes the argument thesis. In this paper, we tackle these tasks in the context of complex arguments on a diverse set of topics. In particular, our dataset consists of manually curated argument trees for 741 controversial topics covering 95,312 unique claims; lines of argument are generally of depth 2 to 6. We find that as the distance between a pair of claims increases along the argument path, determining the relative specificity of a pair of claims becomes easier and determining their relative stance becomes harder. |
| Tasks | |
| Published | 2019-06-26 |
| URL | https://arxiv.org/abs/1906.11313v1 |
| https://arxiv.org/pdf/1906.11313v1.pdf | |
| PWC | https://paperswithcode.com/paper/determining-relative-argument-specificity-and |
| Repo | |
| Framework | |
CLEVR-Ref+: Diagnosing Visual Reasoning with Referring Expressions
| Title | CLEVR-Ref+: Diagnosing Visual Reasoning with Referring Expressions |
| Authors | Runtao Liu, Chenxi Liu, Yutong Bai, Alan Yuille |
| Abstract | Referring object detection and referring image segmentation are important tasks that require joint understanding of visual information and natural language. Yet there has been evidence that current benchmark datasets suffer from bias, and current state-of-the-art models cannot be easily evaluated on their intermediate reasoning process. To address these issues and complement similar efforts in visual question answering, we build CLEVR-Ref+, a synthetic diagnostic dataset for referring expression comprehension. The precise locations and attributes of the objects are readily available, and the referring expressions are automatically associated with functional programs. The synthetic nature allows control over dataset bias (through sampling strategy), and the modular programs enable intermediate reasoning ground truth without human annotators. In addition to evaluating several state-of-the-art models on CLEVR-Ref+, we also propose IEP-Ref, a module network approach that significantly outperforms other models on our dataset. In particular, we present two interesting and important findings using IEP-Ref: (1) the module trained to transform feature maps into segmentation masks can be attached to any intermediate module to reveal the entire reasoning process step-by-step; (2) even if all training data has at least one object referred, IEP-Ref can correctly predict no-foreground when presented with false-premise referring expressions. To the best of our knowledge, this is the first direct and quantitative proof that neural modules behave in the way they are intended. |
| Tasks | Object Detection, Question Answering, Semantic Segmentation, Visual Question Answering, Visual Reasoning |
| Published | 2019-01-03 |
| URL | http://arxiv.org/abs/1901.00850v2 |
| http://arxiv.org/pdf/1901.00850v2.pdf | |
| PWC | https://paperswithcode.com/paper/clevr-ref-diagnosing-visual-reasoning-with |
| Repo | |
| Framework | |
OCGAN: One-class Novelty Detection Using GANs with Constrained Latent Representations
| Title | OCGAN: One-class Novelty Detection Using GANs with Constrained Latent Representations |
| Authors | Pramuditha Perera, Ramesh Nallapati, Bing Xiang |
| Abstract | We present a novel model called OCGAN for the classical problem of one-class novelty detection, where, given a set of examples from a particular class, the goal is to determine if a query example is from the same class. Our solution is based on learning latent representations of in-class examples using a denoising auto-encoder network. The key contribution of our work is our proposal to explicitly constrain the latent space to exclusively represent the given class. In order to accomplish this goal, firstly, we force the latent space to have bounded support by introducing a tanh activation in the encoder’s output layer. Secondly, using a discriminator in the latent space that is trained adversarially, we ensure that encoded representations of in-class examples resemble uniform random samples drawn from the same bounded space. Thirdly, using a second adversarial discriminator in the input space, we ensure all randomly drawn latent samples generate examples that look real. Finally, we introduce a gradient-descent based sampling technique that explores points in the latent space that generate potential out-of-class examples, which are fed back to the network to further train it to generate in-class examples from those points. The effectiveness of the proposed method is measured across four publicly available datasets using two one-class novelty detection protocols where we achieve state-of-the-art results. |
| Tasks | Denoising |
| Published | 2019-03-20 |
| URL | http://arxiv.org/abs/1903.08550v1 |
| http://arxiv.org/pdf/1903.08550v1.pdf | |
| PWC | https://paperswithcode.com/paper/ocgan-one-class-novelty-detection-using-gans |
| Repo | |
| Framework | |
Identification of gatekeeper diseases on the way to cardiovascular mortality
| Title | Identification of gatekeeper diseases on the way to cardiovascular mortality |
| Authors | Nils Haug, Stefan Thurner, Alexandra Kautzky-Willer, Michael Gyimesi, Peter Klimek |
| Abstract | Multimorbidity, the co-occurrence of two or more chronic diseases such as diabetes, obesity or cardiovascular diseases in one patient, is a frequent phenomenon. To make care more efficient, it is of relevance to understand how different diseases condition each other over the life time of a patient. However, most of our current knowledge on such patient careers is either confined to narrow time spans or specific (sets of) diseases. Here, we present a population-wide analysis of long-term patient trajectories by clustering them according to their disease history observed over 17 years. When patients acquire new diseases, their cluster assignment might change. A health trajectory can then be described by a temporal sequence of disease clusters. From the transitions between clusters we construct an age-dependent multilayer network of disease clusters. Random walks on this multilayer network provide a more precise model for the time evolution of multimorbid health states when compared to models that cluster patients based on single diseases. Our results can be used to identify decisive events that potentially determine the future disease trajectory of a patient. We find that for elderly patients the cluster network consists of regions of low, medium and high in-hospital mortality. Diagnoses of diabetes and hypertension are found to strongly increase the likelihood for patients to subsequently move into the high-mortality region later in life. |
| Tasks | |
| Published | 2019-08-02 |
| URL | https://arxiv.org/abs/1908.00920v1 |
| https://arxiv.org/pdf/1908.00920v1.pdf | |
| PWC | https://paperswithcode.com/paper/identification-of-gatekeeper-diseases-on-the |
| Repo | |
| Framework | |
Supervised Learning Approach to Approximate Nearest Neighbor Search
| Title | Supervised Learning Approach to Approximate Nearest Neighbor Search |
| Authors | Ville Hyvönen, Elias Jääsaari, Teemu Roos |
| Abstract | Approximate nearest neighbor search is a classic algorithmic problem where the goal is to design an efficient index structure for fast approximate nearest neighbor queries. We show that it can be framed as a classification problem and solved by training a suitable multi-label classifier and using it as an index. Compared to the existing algorithms, this supervised learning approach has several advantages: it enables adapting an index to the query distribution when the query distribution and the corpus distribution differ; it allows using training sets larger than the corpus; and in principle it enables using any multi-label classifier for approximate nearest neighbor search. We demonstrate these advantages on multiple synthetic and real-world data sets by using a random forest and an ensemble of random projection trees as the base classifiers. |
| Tasks | |
| Published | 2019-10-18 |
| URL | https://arxiv.org/abs/1910.08322v1 |
| https://arxiv.org/pdf/1910.08322v1.pdf | |
| PWC | https://paperswithcode.com/paper/supervised-learning-approach-to-approximate |
| Repo | |
| Framework | |
Distributional Reward Decomposition for Reinforcement Learning
| Title | Distributional Reward Decomposition for Reinforcement Learning |
| Authors | Zichuan Lin, Li Zhao, Derek Yang, Tao Qin, Guangwen Yang, Tie-Yan Liu |
| Abstract | Many reinforcement learning (RL) tasks have specific properties that can be leveraged to modify existing RL algorithms to adapt to those tasks and further improve performance, and a general class of such properties is the multiple reward channel. In those environments the full reward can be decomposed into sub-rewards obtained from different channels. Existing work on reward decomposition either requires prior knowledge of the environment to decompose the full reward, or decomposes reward without prior knowledge but with degraded performance. In this paper, we propose Distributional Reward Decomposition for Reinforcement Learning (DRDRL), a novel reward decomposition algorithm which captures the multiple reward channel structure under distributional setting. Empirically, our method captures the multi-channel structure and discovers meaningful reward decomposition, without any requirements on prior knowledge. Consequently, our agent achieves better performance than existing methods on environments with multiple reward channels. |
| Tasks | |
| Published | 2019-11-06 |
| URL | https://arxiv.org/abs/1911.02166v1 |
| https://arxiv.org/pdf/1911.02166v1.pdf | |
| PWC | https://paperswithcode.com/paper/distributional-reward-decomposition-for |
| Repo | |
| Framework | |
Object as Hotspots: An Anchor-Free 3D Object Detection Approach via Firing of Hotspots
| Title | Object as Hotspots: An Anchor-Free 3D Object Detection Approach via Firing of Hotspots |
| Authors | Qi Chen, Lin Sun, Zhixin Wang, Kui Jia, Alan Yuille |
| Abstract | Accurate 3D object detection in LiDAR based point clouds suffers from the challenges of data sparsity and irregularities. Existing methods strive to organize the points regularly, e.g. voxelize, pass them through a designed 2D/3D neural network, and then define object-level anchors that predict offsets of 3D bounding boxes using collective evidence from all the points on the objects of interest. Converse to the state-of-the-art anchor-based methods, based on the very same nature of data sparsity and irregularities, we observe that even points on an isolated object part are informative about position and orientation of the object. We thus argue in this paper for an approach opposite to existing methods using object-level anchors. Technically, we propose to represent an object as a collection of point cliques; one can intuitively think of these point cliques as hotspots, giving rise to the representation of Object as Hotspots (OHS). Based on OHS, we propose a Hotspot Network (HotSpotNet) that performs 3D object detection via firing of hotspots without setting the predefined bounding boxes. A distinctive feature of HotSpotNet is that it makes predictions directly from individual hotspots, and final results are obtained by aggregating these hotspot predictions. Experiments on the KITTI benchmark show the efficacy of our proposed OHS representation. Our one-stage, anchor-free HotSpotNet beats all other one-stage detectors by at least 2% on cars , cyclists and pedestrian for all difficulty levels. Notably, our proposed method performs better on small and difficult objects and we rank the first among all the submitted methods on pedestrian of KITTI test set. |
| Tasks | 3D Object Detection, Object Detection |
| Published | 2019-12-30 |
| URL | https://arxiv.org/abs/1912.12791v1 |
| https://arxiv.org/pdf/1912.12791v1.pdf | |
| PWC | https://paperswithcode.com/paper/object-as-hotspots-an-anchor-free-3d-object |
| Repo | |
| Framework | |
A new method for determining the filled point of the tooth by Bit-Plane Algorithm
| Title | A new method for determining the filled point of the tooth by Bit-Plane Algorithm |
| Authors | Zahra Alidousti, Maryam Taghizadeh Dehkordi |
| Abstract | Up to now, researchers have applied segmentation techniques in their studies on teeth images, with construction on tooth root length and depth. In this paper, a new approach to the exact identification of the filled points of the tooth is proposed. In this method, the filled teeth are detection by applying the Bit-Plane algorithm on the OPG images. The novelty of the proposed method is that we can use it in medicine for the detection of dental filling and we calculate and present the area of the filled points which may help dentists to assess the filled point of the tooth. The experimental results, confirmed by the dentists, clearly indicate that this method is able to separate the filled points from the rest of healthy teeth completely. |
| Tasks | |
| Published | 2019-07-05 |
| URL | https://arxiv.org/abs/1907.02873v1 |
| https://arxiv.org/pdf/1907.02873v1.pdf | |
| PWC | https://paperswithcode.com/paper/a-new-method-for-determining-the-filled-point |
| Repo | |
| Framework | |
Unsupervised Progressive Learning and the STAM Architecture
| Title | Unsupervised Progressive Learning and the STAM Architecture |
| Authors | James Smith, Seth Baer, Cameron Taylor, Constantine Dovrolis |
| Abstract | We first pose the Unsupervised Progressive Learning (UPL) problem: learning salient representations from a non-stationary stream of unlabeled data in which the number of object classes increases with time. To solve the UPL problem we propose an architecture that involves a module called Self-Taught Associative Memory (STAM). Layered hierarchies of STAM modules learn based on a combination of online clustering, novelty detection, forgetting outliers, and storing only prototypical representations rather than specific examples. We evaluate STAM representations using clustering and classification tasks, relying on limited labeled data for the latter. Even though there are no prior approaches that are directly applicable to the UPL problem, we compare the STAM architecture to a couple of unsupervised and self-supervised deep learning approaches adapted in the UPL context. |
| Tasks | |
| Published | 2019-04-03 |
| URL | https://arxiv.org/abs/1904.02021v4 |
| https://arxiv.org/pdf/1904.02021v4.pdf | |
| PWC | https://paperswithcode.com/paper/unsupervised-continual-learning-and-self |
| Repo | |
| Framework | |
Adversarial camera stickers: A physical camera-based attack on deep learning systems
| Title | Adversarial camera stickers: A physical camera-based attack on deep learning systems |
| Authors | Juncheng Li, Frank R. Schmidt, J. Zico Kolter |
| Abstract | Recent work has documented the susceptibility of deep learning systems to adversarial examples, but most such attacks directly manipulate the digital input to a classifier. Although a smaller line of work considers physical adversarial attacks, in all cases these involve manipulating the object of interest, e.g., putting a physical sticker on an object to misclassify it, or manufacturing an object specifically intended to be misclassified. In this work, we consider an alternative question: is it possible to fool deep classifiers, over all perceived objects of a certain type, by physically manipulating the camera itself? We show that by placing a carefully crafted and mainly-translucent sticker over the lens of a camera, one can create universal perturbations of the observed images that are inconspicuous, yet misclassify target objects as a different (targeted) class. To accomplish this, we propose an iterative procedure for both updating the attack perturbation (to make it adversarial for a given classifier), and the threat model itself (to ensure it is physically realizable). For example, we show that we can achieve physically-realizable attacks that fool ImageNet classifiers in a targeted fashion 49.6% of the time. This presents a new class of physically-realizable threat models to consider in the context of adversarially robust machine learning. Our demo video can be viewed at: https://youtu.be/wUVmL33Fx54 |
| Tasks | |
| Published | 2019-03-21 |
| URL | https://arxiv.org/abs/1904.00759v4 |
| https://arxiv.org/pdf/1904.00759v4.pdf | |
| PWC | https://paperswithcode.com/paper/adversarial-camera-stickers-a-physical-camera |
| Repo | |
| Framework | |
Self-Attentive Document Interaction Networks for Permutation Equivariant Ranking
| Title | Self-Attentive Document Interaction Networks for Permutation Equivariant Ranking |
| Authors | Rama Kumar Pasumarthi, Xuanhui Wang, Michael Bendersky, Marc Najork |
| Abstract | How to leverage cross-document interactions to improve ranking performance is an important topic in information retrieval (IR) research. However, this topic has not been well-studied in the learning-to-rank setting and most of the existing work still treats each document independently while scoring. The recent development of deep learning shows strength in modeling complex relationships across sequences and sets. It thus motivates us to study how to leverage cross-document interactions for learning-to-rank in the deep learning framework. In this paper, we formally define the permutation-equivariance requirement for a scoring function that captures cross-document interactions. We then propose a self-attention based document interaction network and show that it satisfies the permutation-equivariant requirement, and can generate scores for document sets of varying sizes. Our proposed methods can automatically learn to capture document interactions without any auxiliary information, and can scale across large document sets. We conduct experiments on three ranking datasets: the benchmark Web30k, a Gmail search, and a Google Drive Quick Access dataset. Experimental results show that our proposed methods are both more effective and efficient than baselines. |
| Tasks | Information Retrieval, Learning-To-Rank |
| Published | 2019-10-21 |
| URL | https://arxiv.org/abs/1910.09676v2 |
| https://arxiv.org/pdf/1910.09676v2.pdf | |
| PWC | https://paperswithcode.com/paper/self-attentive-document-interaction-networks |
| Repo | |
| Framework | |
Optimal Clustering Framework for Hyperspectral Band Selection
| Title | Optimal Clustering Framework for Hyperspectral Band Selection |
| Authors | Qi Wang, Fahong Zhang, Xuelong Li |
| Abstract | Band selection, by choosing a set of representative bands in hyperspectral image (HSI), is an effective method to reduce the redundant information without compromising the original contents. Recently, various unsupervised band selection methods have been proposed, but most of them are based on approximation algorithms which can only obtain suboptimal solutions toward a specific objective function. This paper focuses on clustering-based band selection, and proposes a new framework to solve the above dilemma, claiming the following contributions: 1) An optimal clustering framework (OCF), which can obtain the optimal clustering result for a particular form of objective function under a reasonable constraint. 2) A rank on clusters strategy (RCS), which provides an effective criterion to select bands on existing clustering structure. 3) An automatic method to determine the number of the required bands, which can better evaluate the distinctive information produced by certain number of bands. In experiments, the proposed algorithm is compared to some state-of-the-art competitors. According to the experimental results, the proposed algorithm is robust and significantly outperform the other methods on various data sets. |
| Tasks | |
| Published | 2019-04-30 |
| URL | http://arxiv.org/abs/1904.13036v1 |
| http://arxiv.org/pdf/1904.13036v1.pdf | |
| PWC | https://paperswithcode.com/paper/optimal-clustering-framework-for |
| Repo | |
| Framework | |
Veridical Data Science
| Title | Veridical Data Science |
| Authors | Bin Yu, Karl Kumbier |
| Abstract | Building and expanding on principles of statistics, machine learning, and scientific inquiry, we propose the predictability, computability, and stability (PCS) framework for veridical data science. Our framework, comprised of both a workflow and documentation, aims to provide responsible, reliable, reproducible, and transparent results across the entire data science life cycle. The PCS workflow uses predictability as a reality check and considers the importance of computation in data collection/storage and algorithm design. It augments predictability and computability with an overarching stability principle for the data science life cycle. Stability expands on statistical uncertainty considerations to assess how human judgment calls impact data results through data and model/algorithm perturbations. Moreover, we develop inference procedures that build on PCS, namely PCS perturbation intervals and PCS hypothesis testing, to investigate the stability of data results relative to problem formulation, data cleaning, modeling decisions, and interpretations. We illustrate PCS inference through neuroscience and genomics projects of our own and others and compare it to existing methods in high dimensional, sparse linear model simulations. Over a wide range of misspecified simulation models, PCS inference demonstrates favorable performance in terms of ROC curves. Finally, we propose PCS documentation based on R Markdown or Jupyter Notebook, with publicly available, reproducible codes and narratives to back up human choices made throughout an analysis. The PCS workflow and documentation are demonstrated in a genomics case study available on Zenodo. |
| Tasks | |
| Published | 2019-01-23 |
| URL | https://arxiv.org/abs/1901.08152v5 |
| https://arxiv.org/pdf/1901.08152v5.pdf | |
| PWC | https://paperswithcode.com/paper/three-principles-of-data-science |
| Repo | |
| Framework | |
How Does Learning Rate Decay Help Modern Neural Networks?
| Title | How Does Learning Rate Decay Help Modern Neural Networks? |
| Authors | Kaichao You, Mingsheng Long, Jianmin Wang, Michael I. Jordan |
| Abstract | Learning rate decay (lrDecay) is a \emph{de facto} technique for training modern neural networks. It starts with a large learning rate and then decays it multiple times. It is empirically observed to help both optimization and generalization. Common beliefs in how lrDecay works come from the optimization analysis of (Stochastic) Gradient Descent: 1) an initially large learning rate accelerates training or helps the network escape spurious local minima; 2) decaying the learning rate helps the network converge to a local minimum and avoid oscillation. Despite the popularity of these common beliefs, experiments suggest that they are insufficient in explaining the general effectiveness of lrDecay in training modern neural networks that are deep, wide, and nonconvex. We provide another novel explanation: an initially large learning rate suppresses the network from memorizing noisy data while decaying the learning rate improves the learning of complex patterns. The proposed explanation is validated on a carefully-constructed dataset with tractable pattern complexity. And its implication, that additional patterns learned in later stages of lrDecay are more complex and thus less transferable, is justified in real-world datasets. We believe that this alternative explanation will shed light into the design of better training strategies for modern neural networks. |
| Tasks | |
| Published | 2019-08-05 |
| URL | https://arxiv.org/abs/1908.01878v2 |
| https://arxiv.org/pdf/1908.01878v2.pdf | |
| PWC | https://paperswithcode.com/paper/learning-stages-phenomenon-root-cause |
| Repo | |
| Framework | |
