Search

Transfer learning for deep neural networks is the process of first training a base network on a source dataset, and then transferring the learned features (the network's weights) to a second network to be trained on a target dataset. This idea has been shown to improve deep neural network's generalization capabilities in many computer vision tasks such as image recognition and object localization. Apart from these applications, deep Convolutional Neural Networks (CNNs) have also recently gained popularity in the Time Series Classification (TSC) community. However, unlike for image recognition problems, transfer learning techniques have not yet been investigated thoroughly for the TSC task. This is surprising as the accuracy of deep learning models for TSC could potentially be improved if the model is fine-tuned from a pre-trained neural network instead of training it from scratch. In this paper, we fill this gap by investigating how to transfer deep CNNs for the TSC task. To evaluate the potential of transfer learning, we performed extensive experiments using the UCR archive which is the largest publicly available TSC benchmark containing 85 datasets. For each dataset in the archive, we pre-trained a model and then fine-tuned it on the other datasets resulting in 7140 different deep neural networks. These experiments revealed that transfer learning can improve or degrade the model's predictions depending on the dataset used for transfer. Therefore, in an effort to predict the best source dataset for a given target dataset, we propose a new method relying on Dynamic Time Warping to measure inter-datasets similarities. We describe how our method can guide the transfer to choose the best source dataset leading to an improvement in accuracy on 71 out of 85 datasets., Comment: Accepted at IEEE International Conference on Big Data 2018

An important part of Digital Pathology is the analysis of multiple digitised whole slide images from differently stained tissue sections. It is common practice to mount consecutive sections containing corresponding microscopic structures on glass slides, and to stain them differently to highlight specific tissue components. These multiple staining modalities result in very different images but include a significant amount of consistent image information. Deep learning approaches have recently been proposed to analyse these images in order to automatically identify objects of interest for pathologists. These supervised approaches require a vast amount of annotations, which are difficult and expensive to acquire---a problem that is multiplied with multiple stainings. This article presents several training strategies that make progress towards stain invariant networks. By training the network on one commonly used staining modality and applying it to images that include corresponding but differently stained tissue structures, the presented unsupervised strategies demonstrate significant improvements over standard training strategies.

Although spin injection at room temperature in an IrMn metallic antiferromagnet strongly depends on the transport regime, and is more efficient in the case of magnonic transport, in this article, we present experimental data demonstrating that the enhanced efficiency of spin injection caused by spin fluctuations near the ordering temperature can be as efficient for the electronic and magnonic transport regimes. By selecting representative interacting environments, we also demonstrated that the amplification of spin injection near the ordering temperature of the IrMn antiferromagnet is independent of exchange coupling with an adjacent NiFe ferromagnet. In addition, our findings confirm that the spin current carried by magnons penetrates deeper than that transported by conduction electrons in IrMn. Finally, our data indicates that the value of the ordering temperature for the IrMn antiferromagnet is not significantly affected by either the electronic or magnonic nature of the spin current probe, or by exchange coupling.

Time Series Classification (TSC) is an important and challenging problem in data mining. With the increase of time series data availability, hundreds of TSC algorithms have been proposed. Among these methods, only a few have considered Deep Neural Networks (DNNs) to perform this task. This is surprising as deep learning has seen very successful applications in the last years. DNNs have indeed revolutionized the field of computer vision especially with the advent of novel deeper architectures such as Residual and Convolutional Neural Networks. Apart from images, sequential data such as text and audio can also be processed with DNNs to reach state-of-the-art performance for document classification and speech recognition. In this article, we study the current state-of-the-art performance of deep learning algorithms for TSC by presenting an empirical study of the most recent DNN architectures for TSC. We give an overview of the most successful deep learning applications in various time series domains under a unified taxonomy of DNNs for TSC. We also provide an open source deep learning framework to the TSC community where we implemented each of the compared approaches and evaluated them on a univariate TSC benchmark (the UCR/UEA archive) and 12 multivariate time series datasets. By training 8,730 deep learning models on 97 time series datasets, we propose the most exhaustive study of DNNs for TSC to date., Comment: Accepted at Data Mining and Knowledge Discovery

Data augmentation in deep neural networks is the process of generating artificial data in order to reduce the variance of the classifier with the goal to reduce the number of errors. This idea has been shown to improve deep neural network's generalization capabilities in many computer vision tasks such as image recognition and object localization. Apart from these applications, deep Convolutional Neural Networks (CNNs) have also recently gained popularity in the Time Series Classification (TSC) community. However, unlike in image recognition problems, data augmentation techniques have not yet been investigated thoroughly for the TSC task. This is surprising as the accuracy of deep learning models for TSC could potentially be improved, especially for small datasets that exhibit overfitting, when a data augmentation method is adopted. In this paper, we fill this gap by investigating the application of a recently proposed data augmentation technique based on the Dynamic Time Warping distance, for a deep learning model for TSC. To evaluate the potential of augmenting the training set, we performed extensive experiments using the UCR TSC benchmark. Our preliminary experiments reveal that data augmentation can drastically increase deep CNN's accuracy on some datasets and significantly improve the deep model's accuracy when the method is used in an ensemble approach., Comment: Accepted at AALTD'18 workshop in ECML/PKDD

Surgical data science is a scientific discipline with the objective of improving the quality of interventional healthcare and its value through capturing, organization, analysis, and modeling of data. The goal of the 1st workshop on Surgical Data Science was to bring together researchers working on diverse topics in surgical data science in order to discuss existing challenges, potential standards and new research directions in the field. Inspired by current open space and think tank formats, it was organized in June 2016 in Heidelberg. While the first day of the workshop, which was dominated by interactive sessions, was open to the public, the second day was reserved for a board meeting on which the information gathered on the public day was processed by (1) discussing remaining open issues, (2) deriving a joint definition for surgical data science and (3) proposing potential strategies for advancing the field. This document summarizes the key findings., Comment: 29 pages

The need for automatic surgical skills assessment is increasing, especially because manual feedback from senior surgeons observing junior surgeons is prone to subjectivity and time consuming. Thus, automating surgical skills evaluation is a very important step towards improving surgical practice. In this paper, we designed a Convolutional Neural Network (CNN) to evaluate surgeon skills by extracting patterns in the surgeon motions performed in robotic surgery. The proposed method is validated on the JIGSAWS dataset and achieved very competitive results with 100% accuracy on the suturing and needle passing tasks. While we leveraged from the CNNs efficiency, we also managed to mitigate its black-box effect using class activation map. This feature allows our method to automatically highlight which parts of the surgical task influenced the skill prediction and can be used to explain the classification and to provide personalized feedback to the trainee., Comment: Accepted at MICCAI 2018

The PHENIX collaboration presents first measurements of low-momentum ($0.4$1\,GeV/$c$) direct-photon yield $dN_{\gamma}^{\rm dir}/d\eta$ is a smooth function of $dN_{\rm ch}/d\eta$ and can be well described as proportional to $(dN_{\rm ch}/d\eta)^\alpha$ with $\alpha{\approx}1.25$. This scaling behavior holds for a wide range of beam energies at the Relativistic Heavy Ion Collider and the Large Hadron Collider, for centrality selected samples, as well as for different, $A$$+$$A$ collision systems. At a given beam energy the scaling also holds for high $p_T$ ($>5$\,GeV/$c$) but when results from different collision energies are compared, an additional $\sqrt{s_{_{NN}}}$-dependent multiplicative factor is needed to describe the integrated-direct-photon yield., Comment: 673 authors from 82 institutions, 10 pages, 4 figures. v2 is version accepted for publication in Physical Review Letters. Plain text data tables for the points plotted in figures for this and previous PHENIX publications are (or will be) publicly available at http://www.phenix.bnl.gov/papers.html

Decentralized crypto-currencies based on the blockchain architecture under-utilize available network bandwidth, making them unable to scale to thousands of transactions per second. We define the Blockclique architecture, that addresses this limitation by sharding transactions in a block graph with a fixed number of threads. The architecture allows the creation of intrinsically compatible blocks in parallel, where each block references one previous block of each thread. The consistency of the Blockclique protocol is formally established in presence of attackers. An experimental evaluation of the architecture's performance in large realistic networks demonstrates an efficient use of available bandwidth and a throughput of thousands of transactions per second.

Intrinsically motivated goal exploration algorithms enable machines to discover repertoires of policies that produce a diversity of effects in complex environments. These exploration algorithms have been shown to allow real world robots to acquire skills such as tool use in high-dimensional continuous state and action spaces. However, they have so far assumed that self-generated goals are sampled in a specifically engineered feature space, limiting their autonomy. In this work, we propose to use deep representation learning algorithms to learn an adequate goal space. This is a developmental 2-stage approach: first, in a perceptual learning stage, deep learning algorithms use passive raw sensor observations of world changes to learn a corresponding latent space; then goal exploration happens in a second stage by sampling goals in this latent space. We present experiments where a simulated robot arm interacts with an object, and we show that exploration algorithms using such learned representations can match the performance obtained using engineered representations.

We propose two general and falsifiable hypotheses about expectations on generalization error when learning in the context of concept drift. One posits that as drift rate increases, the forgetting rate that minimizes generalization error will also increase and vice versa. The other posits that as a learner's forgetting rate increases, the bias/variance profile that minimizes generalization error will have lower variance and vice versa. These hypotheses lead to the concept of the sweet path, a path through the 3-d space of alternative drift rates, forgetting rates and bias/variance profiles on which generalization error will be minimized, such that slow drift is coupled with low forgetting and low bias, while rapid drift is coupled with fast forgetting and low variance. We present experiments that support the existence of such a sweet path. We also demonstrate that simple learners that select appropriate forgetting rates and bias/variance profiles are highly competitive with the state-of-the-art in incremental learners for concept drift on real-world drift problems.