Javier Sánchez Galán, Carlos Quintero, F. Merchan, A. Cornejo
{"title":"Using Deep Convolutional Networks for the Automatic Recognition of MacroInvertebrate in Rivers and Affluents in Panama","authors":"Javier Sánchez Galán, Carlos Quintero, F. Merchan, A. Cornejo","doi":"10.52591/lxai201812037","DOIUrl":"https://doi.org/10.52591/lxai201812037","url":null,"abstract":"The determination of water quality in rivers is a very important task for ensuring the well-being of any population. Moreover, It is very laborious task, involving a combination of chemical and microbiological analysis techniques. A low cost alternative to achieve this same goal is the use of biomonitoring or recognition of macroinvertebrate organisms which presence or absence can be used to determine the water quality. Biomonitoring is often used in rural communities that use of natural streams for consumption and daily life. In those cases community members are often the ones obtaining the quality reports. Although it is an efficient method it has the disadvantage that it requires biological training both for the capture, and identification of macroinvertebrate present in water and watersheds. This work focuses on providing a tool to automate the recognition of freshwater macroinvertebrate images based on deep convolutional networks using the Python programming language, and the Tensorflow and Keras libraries.","PeriodicalId":255707,"journal":{"name":"LatinX in AI at Neural Information Processing Systems Conference 2018","volume":"385 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122353154","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Dehua Chen, Tiago Pimentel, Adriano Veloso, N. Ziviani
{"title":"Denoising Node Embeddings","authors":"Dehua Chen, Tiago Pimentel, Adriano Veloso, N. Ziviani","doi":"10.52591/lxai201812033","DOIUrl":"https://doi.org/10.52591/lxai201812033","url":null,"abstract":"Node embedding (NE) algorithms capture features of graph’s nodes and represent them in a low dimensional vector space. Graphs are inherently noisy structures, which might reduce the learned representations quality. We propose a novel approach using denoising autoencoders to reduce noise in the learned representation of nodes. Experiments with three state-of-the-art NE algorithms show that our approach effectively reduces noise in a link prediction task.","PeriodicalId":255707,"journal":{"name":"LatinX in AI at Neural Information Processing Systems Conference 2018","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124478130","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Topological Data Analysis to Identify Subgroups of Type-2 Diabetes Mellitus","authors":"Juan Vallarta, D. Prieto-Merino","doi":"10.52591/lxai2018120317","DOIUrl":"https://doi.org/10.52591/lxai2018120317","url":null,"abstract":"Introduction There are several factors that contribute to the development of Type-2 Diabetes Mellitus (T2DM) (1); some evidence suggests that it is not only influenced by a deficiency in the pancreatic functions, but from a more complex path-way of the disease (2,3). Unsupervised machine learning algorithms have been applied to gain a better understanding of the T2DM (4,5,6). Topological Data Analysis (TDA) has been recently implemented in the analysis of health datasets (7,8,9,10) to get insight from large and complex sets of data by studying its shape. TDA permits to find clusters of data that other unsupervised methods cannot identify (11). TDA has been applied to identify subtypes of T2DM patients in the American population (12). There is not much information regarding subtypes of this disease and the implementation of TDA. The aim of this research was to perform a TDA in T2DM English patients using linked datasets to identify clusters of the disease.","PeriodicalId":255707,"journal":{"name":"LatinX in AI at Neural Information Processing Systems Conference 2018","volume":"240 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131606884","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Formula 1 rankings prediction by Neural Networks","authors":"D. Patino, Yván Túpac","doi":"10.52591/lxai201812034","DOIUrl":"https://doi.org/10.52591/lxai201812034","url":null,"abstract":"","PeriodicalId":255707,"journal":{"name":"LatinX in AI at Neural Information Processing Systems Conference 2018","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124140734","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Hover: A Wearable Object Identification System for Audio Augmented Reality Interactions","authors":"Pedro Colon Hernandez, Michael Bove","doi":"10.52591/lxai2018120313","DOIUrl":"https://doi.org/10.52591/lxai2018120313","url":null,"abstract":"We design, implement, and test a wearable object identification system which allows users to \"hover\" their hands over objects of interest and get access to contextual information that may be tied to them, through an intelligent personal assistant. The system uses a fusion of sensors to be able to perform the identification of an object under a variety of conditions. Among these sensors there is a camera (operating in the visible and infrared spectrum), a small solid-state radar, and multi-spectral light spectroscopy sensors. Users can interact with contextual information tied to an object through conversations with an intelligent assistant to permit a hands-free, non-obtrusive, and personalized experience. The system explores audio interfacing with augmented reality content without the hassle of phones or head mounted devices.","PeriodicalId":255707,"journal":{"name":"LatinX in AI at Neural Information Processing Systems Conference 2018","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123816025","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"ChaLearn AutoML Challenges","authors":"H. Escalante, Jorge G. Madrid, Eduardo Morales","doi":"10.52591/lxai201812035","DOIUrl":"https://doi.org/10.52591/lxai201812035","url":null,"abstract":"Academic challenges comprise effective means to rapidly advance the state of the art and solve problems of practical relevance. We elaborate on the benefits of challenges for the LatinX community and summarize ChaLearn challenges on Autonomous Machine Learning.","PeriodicalId":255707,"journal":{"name":"LatinX in AI at Neural Information Processing Systems Conference 2018","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131559181","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Predicting Future User Activities with Constrained Non-Linear Models","authors":"Mariano Phielipp, B. Kveton","doi":"10.52591/lxai2018120311","DOIUrl":"https://doi.org/10.52591/lxai2018120311","url":null,"abstract":"Prediction of future user activities from their history, all past activities, is a challenging problem. One reason is that the number of all potential histories grows exponentially with the length of the 4 history. Recently, deep-learning models have been proposed for solving this problem. It is easy to learn a simple predictor of future user activities, by averaging all past activities of the user and then learning an activity classifier from this average representation, for instance by logistic regression. This approach tends to have a high bias, due to using a simple feature representation and model. It is also easy to apply sequence models in deep learning to learn a predictor of future user activities. This approach tends to have a high variance, because of the large number of parameters in the neural network.","PeriodicalId":255707,"journal":{"name":"LatinX in AI at Neural Information Processing Systems Conference 2018","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116594033","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A deep learning approach to sign language recognition using stacked sparse autoencoders","authors":"Pablo Rivas","doi":"10.52591/lxai2018120322","DOIUrl":"https://doi.org/10.52591/lxai2018120322","url":null,"abstract":"","PeriodicalId":255707,"journal":{"name":"LatinX in AI at Neural Information Processing Systems Conference 2018","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124565912","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Monte Carlo Methods on High Dimensional Data","authors":"Rel Guzman","doi":"10.52591/lxai2018120314","DOIUrl":"https://doi.org/10.52591/lxai2018120314","url":null,"abstract":"Markov Chain Monte Carlo (MCMC) simulation is a family of stochastic algorithms that are commonly used to approximate probability distributions by generating samples. The aim of this proposal is to deal with the problem of doing that job on a large scale because due to the increasing power computational demands of data being tall or wide, a study that combines statistical and engineering expertise can be made in order to achieve hardware-accelerated MCMC inference. In this work, I attempt to advance the theory and practice of approximate MCMC methods by developing a toolbox of distributed MCMC algorithms, and then a new method for dealing with large scale problems will be proposed, or else a framework for choosing the most appropriate method will be established. Papers like [1] provide a comprehensive review of the existing literature regarding methods to tackle big data problems. My idea is to tackle divide and conquer approaches since they can work distributed in several machines or else Graphics Processing Unit (GPUs), so I cover the theory behind these methods; then, exhaustive experimental tests will help me compare and categorize them according to their limitations in wide and tall data by considering the dataset size n, sample dimension d, and number of samples T to produce.","PeriodicalId":255707,"journal":{"name":"LatinX in AI at Neural Information Processing Systems Conference 2018","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116776008","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Challenges in Automatic Meme Generation Using Deep Neural Networks","authors":"Albert Orozco, Iván V. Meza","doi":"10.52591/lxai201812032","DOIUrl":"https://doi.org/10.52591/lxai201812032","url":null,"abstract":"","PeriodicalId":255707,"journal":{"name":"LatinX in AI at Neural Information Processing Systems Conference 2018","volume":"149 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116388602","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}