1,720,973 research outputs found
Monotone Deep Spectrum Kernels
No full textA recent result in the literature states that polynomial and conjunctive features can be hierarchically organized and described by different kernels of increasing expressiveness (or complexity). Additionally, the optimal combination of those kernels through a Multiple Kernel Learning approach produces effective and robust deep kernels. In this paper, we extend this approach to structured data, showing an adaptation of classical spectrum kernels, here named monotone spectrum kernels, reflecting a hierarchical feature space of sub-structures of increasing complexity. Finally, we show that (i) our kernels adaptation does not differ significantly from classical spectrum kernels, and (ii) the optimal combination achieves better results than the single spectrum kernel
Distribution-Based Global Sensitivity Analysis in Hydrology
No full textGlobal sensitivity analysis (GSA) is routinely used in academic setting to quantify the influence of input variability and uncertainty on predictions of a quantity of interest. Practical applications of GSA are hampered by its high computational cost, which arises from the need to run large (e.g., groundwater) models multiple times, and by its reliance on the analysis of variance, which formally requires input parameters to be uncorrelated. The former difficulty can be alleviated by replacing expensive models with inexpensive (e.g., polynomial) surrogates, while adoption of distribution‐based (rather than variance‐based) metrics can, in principle, overcome the latter but at significantly increased computational cost. To make use of distribution‐based GSA feasible for regional‐scale models with a large number of degrees of freedom, we supplement it with a surrogate model built with polynomial chaos expansions with analytically updated coefficients. We demonstrate the computational efficiency of our algorithm on a case study dealing with evaluation of the effects of temperature variability on annual evapotranspiration at the regional scale
Enhancing deep neural networks via multiple kernel learning
No full textDeep neural networks and Multiple Kernel Learning are representation learning methodologies of widespread use and increasing success. While the former aims at learning representations through a hierarchy of features of increasing complexity, the latter provides a principled approach for the combination of base representations. In this paper, we introduce a general framework in which the internal representations computed by a deep neural network are optimally combined by means of Multiple Kernel Learning. The resulting ensemble methodology is instantiated for Multi-layer Perceptrons architectures (both fully trained and with random-weights), and for Convolutional Neural Networks. Experimental results on several benchmark datasets concretely show the advantages and potentialities of the proposed approach
Language processing in the era of deep learning
No full textNatural Language Processing is a branch of artificial in- telligence brimful of intricate, sophisticated, and challenging tasks, such as machine translation, question answering, summarization, and so on. Thanks to the recent advances of deep learning, NLP applications have received an unprecedented boost in performance, generating growing in- terest from the Machine Learning community. However, even if recent techniques are starting to reach excellent performance on various tasks, there are still several problems that need to be solved, such as the compu- tational cost, the reproducibility of results, and the lack of interpretability. In this contribution, we provide a high-level overview of recent advances in NLP, the role of Machine Learning, and current research directions
Exploring the feature space of character-level embeddings
No full textRecently, character-level embeddings have become popular in the Natural Language Processing community. These representations provide a description of a word which depends solely on its inner structure, i.e. the sequence of characters. Convolutional and recurrent neural networks are the undisputed protagonists in this context, and they represent the state of the art for many character-level applications. In this work, we firstly compare different neural architectures against adaptive string kernels in simplified scenarios. Then, we propose a hybrid ensemble that injects structural kernel-based features into a neural architecture, providing an efficient and scalable solution. An all-around experimental assessment has been carried out on several string datasets, including biomedical entity recognition and sentiment analysis
Learning deep kernels in the space of monotone conjunctive polynomials
No full textDot-product kernels is a large family of kernel functions based on dot-product between examples. A recent result states that any dot-product kernel can be decomposed as a non-negative linear combination of homogeneous polynomial kernels of different degrees, and it is possible to learn the coefficients of the combination by exploiting the Multiple Kernel Learning (MKL) paradigm. In this paper it is proved that, under mild conditions, any homogeneous polynomial kernel defined on binary valued data can be decomposed in a parametrized finite linear non-negative combination of monotone conjunctive kernels. MKL has been employed to learn the parameters of the combination. Furthermore, we show that our solution produces a deep kernel whose feature space consists of hierarchically organized features of increasing complexity. We also emphasize the connection between our solution and existing deep kernel learning frameworks. A wide empirical assessment is presented to evaluate the proposed framework, and to compare it against the baselines on several categorical and binary datasets
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
- …
