Paper Group ANR 720
Feature versus Raw Sequence: Deep Learning Comparative Study on Predicting Pre-miRNA. Joint Max Margin and Semantic Features for Continuous Event Detection in Complex Scenes. A New 3D Method to Segment the Lumbar Vertebral Bodies and to Determine Bone Mineral Density and Geometry. Data Decisions and Theoretical Implications when Adversarially Learn …
Feature versus Raw Sequence: Deep Learning Comparative Study on Predicting Pre-miRNA
| Title | Feature versus Raw Sequence: Deep Learning Comparative Study on Predicting Pre-miRNA |
| Authors | Jaya Thomas, Sonia Thomas, Lee Sael |
| Abstract | Should we input known genome sequence features or input sequence itself in deep learning framework? As deep learning more popular in various applications, researchers often come to question whether to generate features or use raw sequences for deep learning. To answer this question, we study the prediction accuracy of precursor miRNA prediction of feature-based deep belief network and sequence-based convolution neural network. Tested on a variant of six-layer convolution neural net and three-layer deep belief network, we find the raw sequence input based convolution neural network model performs similar or slightly better than feature based deep belief networks with best accuracy values of 0.995 and 0.990, respectively. Both the models outperform existing benchmarks models. The results shows us that if provided large enough data, well devised raw sequence based deep learning models can replace feature based deep learning models. However, construction of well behaved deep learning model can be very challenging. In cased features can be easily extracted, feature-based deep learning models may be a better alternative. |
| Tasks | |
| Published | 2017-10-17 |
| URL | http://arxiv.org/abs/1710.06798v1 |
| http://arxiv.org/pdf/1710.06798v1.pdf | |
| PWC | https://paperswithcode.com/paper/feature-versus-raw-sequence-deep-learning |
| Repo | |
| Framework | |
Joint Max Margin and Semantic Features for Continuous Event Detection in Complex Scenes
| Title | Joint Max Margin and Semantic Features for Continuous Event Detection in Complex Scenes |
| Authors | Iman Abbasnejad, Sridha Sridharan, Simon Denman, Clinton Fookes, Simon Lucey |
| Abstract | In this paper the problem of complex event detection in the continuous domain (i.e. events with unknown starting and ending locations) is addressed. Existing event detection methods are limited to features that are extracted from the local spatial or spatio-temporal patches from the videos. However, this makes the model vulnerable to the events with similar concepts e.g. “Open drawer” and “Open cupboard”. In this work, in order to address the aforementioned limitations we present a novel model based on the combination of semantic and temporal features extracted from video frames. We train a max-margin classifier on top of the extracted features in an adaptive framework that is able to detect the events with unknown starting and ending locations. Our model is based on the Bidirectional Region Neural Network and large margin Structural Output SVM. The generality of our model allows it to be simply applied to different labeled and unlabeled datasets. We finally test our algorithm on three challenging datasets, “UCF 101-Action Recognition”, “MPII Cooking Activities” and “Hollywood”, and we report state-of-the-art performance. |
| Tasks | Temporal Action Localization |
| Published | 2017-06-13 |
| URL | http://arxiv.org/abs/1706.04122v1 |
| http://arxiv.org/pdf/1706.04122v1.pdf | |
| PWC | https://paperswithcode.com/paper/joint-max-margin-and-semantic-features-for |
| Repo | |
| Framework | |
A New 3D Method to Segment the Lumbar Vertebral Bodies and to Determine Bone Mineral Density and Geometry
| Title | A New 3D Method to Segment the Lumbar Vertebral Bodies and to Determine Bone Mineral Density and Geometry |
| Authors | Andre Mastmeyer, Klaus Engelke, Sebastian Meller, Willi Kalender |
| Abstract | In this paper we present a new 3D segmentation approach for the vertebrae of the lower thoracic and the lumbar spine in spiral computed tomography datasets. We implemented a multi-step procedure. Its main components are deformable models, volume growing, and morphological operations. The performance analysis that included an evaluation of accuracy using the European Spine Phantom, and of intra-operator precision using clinical CT datasets from 10 patients highlight the potential for clinical use. The intra-operator precision of the segmentation procedure was better than 1% for Bone Mineral Density (BMD) and better than 1.8% for volume. The long-term goal of this work is to enable better fracture prediction and improved patient monitoring in the field of osteoporosis. A true 3D segmentation also enables an accurate measurement of geometrical parameters that can augment the classical measurement of BMD. |
| Tasks | |
| Published | 2017-05-19 |
| URL | http://arxiv.org/abs/1705.07146v1 |
| http://arxiv.org/pdf/1705.07146v1.pdf | |
| PWC | https://paperswithcode.com/paper/a-new-3d-method-to-segment-the-lumbar |
| Repo | |
| Framework | |
Data Decisions and Theoretical Implications when Adversarially Learning Fair Representations
| Title | Data Decisions and Theoretical Implications when Adversarially Learning Fair Representations |
| Authors | Alex Beutel, Jilin Chen, Zhe Zhao, Ed H. Chi |
| Abstract | How can we learn a classifier that is “fair” for a protected or sensitive group, when we do not know if the input to the classifier belongs to the protected group? How can we train such a classifier when data on the protected group is difficult to attain? In many settings, finding out the sensitive input attribute can be prohibitively expensive even during model training, and sometimes impossible during model serving. For example, in recommender systems, if we want to predict if a user will click on a given recommendation, we often do not know many attributes of the user, e.g., race or age, and many attributes of the content are hard to determine, e.g., the language or topic. Thus, it is not feasible to use a different classifier calibrated based on knowledge of the sensitive attribute. Here, we use an adversarial training procedure to remove information about the sensitive attribute from the latent representation learned by a neural network. In particular, we study how the choice of data for the adversarial training effects the resulting fairness properties. We find two interesting results: a small amount of data is needed to train these adversarial models, and the data distribution empirically drives the adversary’s notion of fairness. |
| Tasks | Recommendation Systems |
| Published | 2017-07-01 |
| URL | http://arxiv.org/abs/1707.00075v2 |
| http://arxiv.org/pdf/1707.00075v2.pdf | |
| PWC | https://paperswithcode.com/paper/data-decisions-and-theoretical-implications |
| Repo | |
| Framework | |
Deep Transfer Learning for Error Decoding from Non-Invasive EEG
| Title | Deep Transfer Learning for Error Decoding from Non-Invasive EEG |
| Authors | Martin Völker, Robin T. Schirrmeister, Lukas D. J. Fiederer, Wolfram Burgard, Tonio Ball |
| Abstract | We recorded high-density EEG in a flanker task experiment (31 subjects) and an online BCI control paradigm (4 subjects). On these datasets, we evaluated the use of transfer learning for error decoding with deep convolutional neural networks (deep ConvNets). In comparison with a regularized linear discriminant analysis (rLDA) classifier, ConvNets were significantly better in both intra- and inter-subject decoding, achieving an average accuracy of 84.1 % within subject and 81.7 % on unknown subjects (flanker task). Neither method was, however, able to generalize reliably between paradigms. Visualization of features the ConvNets learned from the data showed plausible patterns of brain activity, revealing both similarities and differences between the different kinds of errors. Our findings indicate that deep learning techniques are useful to infer information about the correctness of action in BCI applications, particularly for the transfer of pre-trained classifiers to new recording sessions or subjects. |
| Tasks | EEG, Transfer Learning |
| Published | 2017-10-25 |
| URL | http://arxiv.org/abs/1710.09139v3 |
| http://arxiv.org/pdf/1710.09139v3.pdf | |
| PWC | https://paperswithcode.com/paper/deep-transfer-learning-for-error-decoding |
| Repo | |
| Framework | |
An Error-Oriented Approach to Word Embedding Pre-Training
| Title | An Error-Oriented Approach to Word Embedding Pre-Training |
| Authors | Youmna Farag, Marek Rei, Ted Briscoe |
| Abstract | We propose a novel word embedding pre-training approach that exploits writing errors in learners’ scripts. We compare our method to previous models that tune the embeddings based on script scores and the discrimination between correct and corrupt word contexts in addition to the generic commonly-used embeddings pre-trained on large corpora. The comparison is achieved by using the aforementioned models to bootstrap a neural network that learns to predict a holistic score for scripts. Furthermore, we investigate augmenting our model with error corrections and monitor the impact on performance. Our results show that our error-oriented approach outperforms other comparable ones which is further demonstrated when training on more data. Additionally, extending the model with corrections provides further performance gains when data sparsity is an issue. |
| Tasks | |
| Published | 2017-07-21 |
| URL | http://arxiv.org/abs/1707.06841v1 |
| http://arxiv.org/pdf/1707.06841v1.pdf | |
| PWC | https://paperswithcode.com/paper/an-error-oriented-approach-to-word-embedding |
| Repo | |
| Framework | |
Incorporating Syntactic Uncertainty in Neural Machine Translation with Forest-to-Sequence Model
| Title | Incorporating Syntactic Uncertainty in Neural Machine Translation with Forest-to-Sequence Model |
| Authors | Poorya Zaremoodi, Gholamreza Haffari |
| Abstract | Incorporating syntactic information in Neural Machine Translation models is a method to compensate their requirement for a large amount of parallel training text, especially for low-resource language pairs. Previous works on using syntactic information provided by (inevitably error-prone) parsers has been promising. In this paper, we propose a forest-to-sequence Attentional Neural Machine Translation model to make use of exponentially many parse trees of the source sentence to compensate for the parser errors. Our method represents the collection of parse trees as a packed forest, and learns a neural attentional transduction model from the forest to the target sentence. Experiments on English to German, Chinese and Persian translation show the superiority of our method over the tree-to-sequence and vanilla sequence-to-sequence neural translation models. |
| Tasks | Machine Translation |
| Published | 2017-11-19 |
| URL | http://arxiv.org/abs/1711.07019v2 |
| http://arxiv.org/pdf/1711.07019v2.pdf | |
| PWC | https://paperswithcode.com/paper/incorporating-syntactic-uncertainty-in-neural-1 |
| Repo | |
| Framework | |
Towards high-throughput 3D insect capture for species discovery and diagnostics
| Title | Towards high-throughput 3D insect capture for species discovery and diagnostics |
| Authors | Chuong Nguyen, Matt Adcock, Stuart Anderson, David Lovell, Nicole Fisher, John La Salle |
| Abstract | Digitisation of natural history collections not only preserves precious information about biological diversity, it also enables us to share, analyse, annotate and compare specimens to gain new insights. High-resolution, full-colour 3D capture of biological specimens yields color and geometry information complementary to other techniques (e.g., 2D capture, electron scanning and micro computed tomography). However 3D colour capture of small specimens is slow for reasons including specimen handling, the narrow depth of field of high magnification optics, and the large number of images required to resolve complex shapes of specimens. In this paper, we outline techniques to accelerate 3D image capture, including using a desktop robotic arm to automate the insect handling process; using a calibrated pan-tilt rig to avoid attaching calibration targets to specimens; using light field cameras to capture images at an extended depth of field in one shot; and using 3D Web and mixed reality tools to facilitate the annotation, distribution and visualisation of 3D digital models. |
| Tasks | Calibration |
| Published | 2017-09-07 |
| URL | http://arxiv.org/abs/1709.02033v1 |
| http://arxiv.org/pdf/1709.02033v1.pdf | |
| PWC | https://paperswithcode.com/paper/towards-high-throughput-3d-insect-capture-for |
| Repo | |
| Framework | |
Dual-Tree Wavelet Scattering Network with Parametric Log Transformation for Object Classification
| Title | Dual-Tree Wavelet Scattering Network with Parametric Log Transformation for Object Classification |
| Authors | Amarjot Singh, Nick Kingsbury |
| Abstract | We introduce a ScatterNet that uses a parametric log transformation with Dual-Tree complex wavelets to extract translation invariant representations from a multi-resolution image. The parametric transformation aids the OLS pruning algorithm by converting the skewed distributions into relatively mean-symmetric distributions while the Dual-Tree wavelets improve the computational efficiency of the network. The proposed network is shown to outperform Mallat’s ScatterNet on two image datasets, both for classification accuracy and computational efficiency. The advantages of the proposed network over other supervised and some unsupervised methods are also presented using experiments performed on different training dataset sizes. |
| Tasks | Object Classification |
| Published | 2017-02-10 |
| URL | http://arxiv.org/abs/1702.03267v1 |
| http://arxiv.org/pdf/1702.03267v1.pdf | |
| PWC | https://paperswithcode.com/paper/dual-tree-wavelet-scattering-network-with |
| Repo | |
| Framework | |
Multi-dimensional Graph Fourier Transform
| Title | Multi-dimensional Graph Fourier Transform |
| Authors | Takashi Kurokawa, Taihei Oki, Hiromichi Nagao |
| Abstract | Many signals on Cartesian product graphs appear in the real world, such as digital images, sensor observation time series, and movie ratings on Netflix. These signals are “multi-dimensional” and have directional characteristics along each factor graph. However, the existing graph Fourier transform does not distinguish these directions, and assigns 1-D spectra to signals on product graphs. Further, these spectra are often multi-valued at some frequencies. Our main result is a multi-dimensional graph Fourier transform that solves such problems associated with the conventional GFT. Using algebraic properties of Cartesian products, the proposed transform rearranges 1-D spectra obtained by the conventional GFT into the multi-dimensional frequency domain, of which each dimension represents a directional frequency along each factor graph. Thus, the multi-dimensional graph Fourier transform enables directional frequency analysis, in addition to frequency analysis with the conventional GFT. Moreover, this rearrangement resolves the multi-valuedness of spectra in some cases. The multi-dimensional graph Fourier transform is a foundation of novel filterings and stationarities that utilize dimensional information of graph signals, which are also discussed in this study. The proposed methods are applicable to a wide variety of data that can be regarded as signals on Cartesian product graphs. This study also notes that multivariate graph signals can be regarded as 2-D univariate graph signals. This correspondence provides natural definitions of the multivariate graph Fourier transform and the multivariate stationarity based on their 2-D univariate versions. |
| Tasks | Time Series |
| Published | 2017-12-21 |
| URL | http://arxiv.org/abs/1712.07811v1 |
| http://arxiv.org/pdf/1712.07811v1.pdf | |
| PWC | https://paperswithcode.com/paper/multi-dimensional-graph-fourier-transform |
| Repo | |
| Framework | |
Stacked Convolutional and Recurrent Neural Networks for Bird Audio Detection
| Title | Stacked Convolutional and Recurrent Neural Networks for Bird Audio Detection |
| Authors | Sharath Adavanne, Konstantinos Drossos, Emre Çakır, Tuomas Virtanen |
| Abstract | This paper studies the detection of bird calls in audio segments using stacked convolutional and recurrent neural networks. Data augmentation by blocks mixing and domain adaptation using a novel method of test mixing are proposed and evaluated in regard to making the method robust to unseen data. The contributions of two kinds of acoustic features (dominant frequency and log mel-band energy) and their combinations are studied in the context of bird audio detection. Our best achieved AUC measure on five cross-validations of the development data is 95.5% and 88.1% on the unseen evaluation data. |
| Tasks | Bird Audio Detection, Data Augmentation, Domain Adaptation |
| Published | 2017-06-07 |
| URL | http://arxiv.org/abs/1706.02047v1 |
| http://arxiv.org/pdf/1706.02047v1.pdf | |
| PWC | https://paperswithcode.com/paper/stacked-convolutional-and-recurrent-neural-1 |
| Repo | |
| Framework | |
A minimax and asymptotically optimal algorithm for stochastic bandits
| Title | A minimax and asymptotically optimal algorithm for stochastic bandits |
| Authors | Pierre Ménard, Aurélien Garivier |
| Abstract | We propose the kl-UCB ++ algorithm for regret minimization in stochastic bandit models with exponential families of distributions. We prove that it is simultaneously asymptotically optimal (in the sense of Lai and Robbins’ lower bound) and minimax optimal. This is the first algorithm proved to enjoy these two properties at the same time. This work thus merges two different lines of research with simple and clear proofs. |
| Tasks | |
| Published | 2017-02-23 |
| URL | http://arxiv.org/abs/1702.07211v2 |
| http://arxiv.org/pdf/1702.07211v2.pdf | |
| PWC | https://paperswithcode.com/paper/a-minimax-and-asymptotically-optimal |
| Repo | |
| Framework | |
Openmv: A Python powered, extensible machine vision camera
| Title | Openmv: A Python powered, extensible machine vision camera |
| Authors | Ibrahim Abdelkader, Yasser El-Sonbaty, Mohamed El-Habrouk |
| Abstract | Advances in semiconductor manufacturing processes and large scale integration keep pushing demanding applications further away from centralized processing, and closer to the edges of the network (i.e. Edge Computing). It has become possible to perform complex in-network image processing using low-power embedded smart cameras, enabling a multitude of new collaborative image processing applications. This paper introduces OpenMV, a new low-power smart camera that lends itself naturally to wireless sensor networks and machine vision applications. The uniqueness of this platform lies in running an embedded Python3 interpreter, allowing its peripherals and machine vision library to be scripted in Python. In addition, its hardware is extensible via modules that augment the platform with new capabilities, such as thermal imaging and networking modules. |
| Tasks | |
| Published | 2017-11-01 |
| URL | http://arxiv.org/abs/1711.10464v1 |
| http://arxiv.org/pdf/1711.10464v1.pdf | |
| PWC | https://paperswithcode.com/paper/openmv-a-python-powered-extensible-machine |
| Repo | |
| Framework | |
Free energy-based reinforcement learning using a quantum processor
| Title | Free energy-based reinforcement learning using a quantum processor |
| Authors | Anna Levit, Daniel Crawford, Navid Ghadermarzy, Jaspreet S. Oberoi, Ehsan Zahedinejad, Pooya Ronagh |
| Abstract | Recent theoretical and experimental results suggest the possibility of using current and near-future quantum hardware in challenging sampling tasks. In this paper, we introduce free energy-based reinforcement learning (FERL) as an application of quantum hardware. We propose a method for processing a quantum annealer’s measured qubit spin configurations in approximating the free energy of a quantum Boltzmann machine (QBM). We then apply this method to perform reinforcement learning on the grid-world problem using the D-Wave 2000Q quantum annealer. The experimental results show that our technique is a promising method for harnessing the power of quantum sampling in reinforcement learning tasks. |
| Tasks | |
| Published | 2017-05-29 |
| URL | http://arxiv.org/abs/1706.00074v1 |
| http://arxiv.org/pdf/1706.00074v1.pdf | |
| PWC | https://paperswithcode.com/paper/free-energy-based-reinforcement-learning |
| Repo | |
| Framework | |
Automated Website Fingerprinting through Deep Learning
| Title | Automated Website Fingerprinting through Deep Learning |
| Authors | Vera Rimmer, Davy Preuveneers, Marc Juarez, Tom Van Goethem, Wouter Joosen |
| Abstract | Several studies have shown that the network traffic that is generated by a visit to a website over Tor reveals information specific to the website through the timing and sizes of network packets. By capturing traffic traces between users and their Tor entry guard, a network eavesdropper can leverage this meta-data to reveal which website Tor users are visiting. The success of such attacks heavily depends on the particular set of traffic features that are used to construct the fingerprint. Typically, these features are manually engineered and, as such, any change introduced to the Tor network can render these carefully constructed features ineffective. In this paper, we show that an adversary can automate the feature engineering process, and thus automatically deanonymize Tor traffic by applying our novel method based on deep learning. We collect a dataset comprised of more than three million network traces, which is the largest dataset of web traffic ever used for website fingerprinting, and find that the performance achieved by our deep learning approaches is comparable to known methods which include various research efforts spanning over multiple years. The obtained success rate exceeds 96% for a closed world of 100 websites and 94% for our biggest closed world of 900 classes. In our open world evaluation, the most performant deep learning model is 2% more accurate than the state-of-the-art attack. Furthermore, we show that the implicit features automatically learned by our approach are far more resilient to dynamic changes of web content over time. We conclude that the ability to automatically construct the most relevant traffic features and perform accurate traffic recognition makes our deep learning based approach an efficient, flexible and robust technique for website fingerprinting. |
| Tasks | Feature Engineering |
| Published | 2017-08-21 |
| URL | http://arxiv.org/abs/1708.06376v2 |
| http://arxiv.org/pdf/1708.06376v2.pdf | |
| PWC | https://paperswithcode.com/paper/automated-website-fingerprinting-through-deep |
| Repo | |
| Framework | |
