作者: 捐助 時(shí)間: 2025-3-22 00:16 作者: 冒煙 時(shí)間: 2025-3-22 02:15
?ngelo Gregório Lovatto,Leliane Nunes de Barros,Denis D. Mauáundles. In Sec. 1 we shall give the basic definitions of the Hermitian analogues of the classical concepts of (Riemannian) metric, connection, and curvature. This is carried out in the context of differentiable C-vector bundles over a differentiable manifold . More specific formulas are obtained in 作者: 注入 時(shí)間: 2025-3-22 07:22
Arnaldo Vitor Barros da Silva,Luis Filipe Alves Pereiraolds. In Sec. 1 we shall present a discussion of the exterior algebra on a Hermitian vector space, introducing the fundamental 2-form and the Hodge *-operator associated with the Hermitian metric. In Sec. 2 we shall discuss and prove the principal results concerning harmonic forms on compact manifol作者: pericardium 時(shí)間: 2025-3-22 10:57
Igor Felipe de Camargo,Rodolfo Stoffel Antunes,Gabriel de O. Ramoss likely to need. The techniques described are presented with due regard for their theoretical basis; but the emphasis is on detailed discussion of the ideas of the differ- ential calculus and on the avoidance of false statements rather than on complete proofs of all results. It is a frequent experi作者: 良心 時(shí)間: 2025-3-22 15:28 作者: 違法事實(shí) 時(shí)間: 2025-3-22 20:02 作者: explicit 時(shí)間: 2025-3-22 23:21
,Least-Squares Linear Dilation-Erosion Regressor Trained Using a?Convex-Concave Procedure,a convex combination of the composition of linear and morphological operators. They yield continuous piecewise linear functions and, thus, are universal approximators. Besides introducing the .-DER model, we formulate their training as a difference of convex (DC) programming problem. Precisely, an .作者: 不公開(kāi) 時(shí)間: 2025-3-23 05:24
,Exploration Versus Exploitation in?Model-Based Reinforcement Learning: An Empirical Study,n used to select a policy that maximizes the objective function. Stochastic Value Gradient (SVG) methods perform the latter step by optimizing some estimate of the value function gradient. Despite showing promising empirical results, many implementations of SVG methods lack rigorous theoretical or e作者: Humble 時(shí)間: 2025-3-23 06:07 作者: tattle 時(shí)間: 2025-3-23 10:00
,Using BERT to?Predict the?Brazilian Stock Market,nt in news with the following approaches: (1) lexicons; or (2) machine learning algorithms. However, both strategies are subject to errors introduced by human factors. While the lexical approach needs a specific dictionary for the financial domain, algorithms based on machine learning need a databas作者: expeditious 時(shí)間: 2025-3-23 16:59
,Human Action Recognition Based on?2D Poses and?Skeleton Joints,e been studied and several researches have emerged using videos as objects of study. One of them is human action recognition, which seeks to identify in a given video what actions the people present in it are performing, whether for recreational purposes or for monitoring public places for people’s 作者: 機(jī)構(gòu) 時(shí)間: 2025-3-23 20:22 作者: HALO 時(shí)間: 2025-3-24 00:37
,When Less May Be More: Exploring Similarity to?Improve Experience Replay,he total accumulated rewards in the long run. COMPER uses temporal difference learning with predicted target values for sets of similar transitions and a new experience replay approach based on two memories of transitions. We present an assessment of two possible neural network architectures for the作者: FLAX 時(shí)間: 2025-3-24 05:32
,Human Identification Based on?Gait and?Soft Biometrics,en approached in many works over the last decades, that resulted in state-of-art methods based on biometric features such as fingerprint, iris and face. Despite the great development in this area, there are still many challenges to overcome, and this present work aims to present an approach to one o作者: alcohol-abuse 時(shí)間: 2025-3-24 09:22 作者: 衰弱的心 時(shí)間: 2025-3-24 10:46
,Combining Neural Networks and?a?Color Classifier for?Fire Detection,ection can bring great improvements to security systems and prevent many losses. Effective fire detection systems help fire situations to have lesser consequences than they could have by signaling as quickly as possible the experts. This paper proposes a combination of modern image classification mo作者: 表狀態(tài) 時(shí)間: 2025-3-24 17:11 作者: 反復(fù)拉緊 時(shí)間: 2025-3-24 21:14
,A Probabilistically-Oriented Analysis of?the?Performance of?ASR Systems for?Brazilian Radios and?TVin the last few years. Several state-of-art results were achieved by open-source end-to-end models, such as the Kaldi toolkit and the Wav2vec 2.0. Alternative commercial tools are also available, including the Google and Microsoft speech to text APIs and the Audimus System of VoiceInteraction. We an作者: 直言不諱 時(shí)間: 2025-3-25 03:13
,A Reward Function Using Image Processing for?a?Deep Reinforcement Learning Approach Applied to?the?ears. Most researches advocate for learning through high-dimensional sensory inputs (i.e. images) as observations in order to simulate scenarios that more closely approach reality. Though, when using these video game environments, the common practice is to provide the agent a reward signal calculate作者: 易受騙 時(shí)間: 2025-3-25 06:45 作者: CRUDE 時(shí)間: 2025-3-25 07:56 作者: 天賦 時(shí)間: 2025-3-25 12:24
Event Detection in Therapy Sessions for Children with Autism,these individuals. ASD has a worldwide prevalence of 1% on children and it affects not only the people with this disorder, but also their family and the surrounding community. In the family circle, individuals on the spectrum require greater support and attention relative to its cognitive capacity, 作者: AWE 時(shí)間: 2025-3-25 19:43
,On Social Consensus Mechanisms for?Federated Learning Aggregation,erated Learning stands out for allowing accurate inference by combining multiple models trained on their data locally. It uses the strategy of local learning, where each node learns from its individual data, and then groups the learned models into a single, unified one, thus preserving data privacy.作者: IRS 時(shí)間: 2025-3-25 20:16
,: A Corpus of?Literature in?Portuguese for?NLP,onal Creativity tasks, such as Natural Language Processing, Automatic Text Generation (ATG), and other similar purposes. We highlight characteristics of the Portuguese section, such as the numbers of documents, authors, sentences and tokens and also how it is structured and formatted. We show how th作者: EXCEL 時(shí)間: 2025-3-26 00:28
,Protein Molecular Function Annotation Based on?Transformer Embeddings,esent and discuss a new approach for protein Molecular Function prediction based on Transformers embeddings. Our method surpassed state-of-the-art classifiers when it used only the amino acid sequence as input and when it employed amino acid sequence and homology search in . and AuPRC metrics.作者: 大氣層 時(shí)間: 2025-3-26 07:29 作者: 偽善 時(shí)間: 2025-3-26 08:31
,Least-Squares Linear Dilation-Erosion Regressor Trained Using a?Convex-Concave Procedure,firm the efficacy of the proposed regressor, outperforming other hybrid morphological models and state-of-the-art approaches such as the multilayer perceptron network and the radial-basis support vector regressor.作者: Lasting 時(shí)間: 2025-3-26 13:01 作者: ANA 時(shí)間: 2025-3-26 19:58 作者: 不知疲倦 時(shí)間: 2025-3-26 23:35
,Self-learning Methodology Based on?Degradation Estimation for?Underwater Image Enhancement,s an enhanced version of the input image. It is highlighted that the proposed algorithm requires only one image as input during the training. The results obtained using our method show its effectiveness of color preservation, color cast reduction, and contrast improvement.作者: RADE 時(shí)間: 2025-3-27 03:37 作者: 充滿人 時(shí)間: 2025-3-27 09:20 作者: Scleroderma 時(shí)間: 2025-3-27 09:59
,: A Corpus of?Literature in?Portuguese for?NLP,e ATG algorithms, which we have previously developed, behave when trained on this corpus, by using a human evaluation protocol where a mixture of automatically generated and . texts is classified, using four criteria: grammaticality, coherence, identification of context, and an adapted Turing test.作者: Decrepit 時(shí)間: 2025-3-27 17:11
0302-9743 2022, which took place in Campinas, Brazil, in November/December 2022.?.The 89 papers presented in the proceedings were carefully reviewed and selected from 225 submissions. The conference deals with theoretical aspects and applications of artificial and computational intelligence..978-3-031-21688-作者: 教育學(xué) 時(shí)間: 2025-3-27 18:49
,A Sequential Recommender System with?Embeddings Based on?GraphSage Aggregators,ing and LSTM. We validated our proposal with the datasets: yoochoose, diginetica, aotm and 30music. The results indicate that, with the Mean aggregator, it is possible to reduce the execution time in all tested scenarios, maintaining the original effectiveness.作者: Sad570 時(shí)間: 2025-3-28 02:00 作者: 浸軟 時(shí)間: 2025-3-28 05:17
,BoVW-CAM: Visual Explanation from?Bag of?Visual Words, techniques for handcrafted features. This fact obscures the comparison between modern CNNs and classical methods for image classification. In this work, we present the BoVW-CAM that indicates the most important image regions for each prediction given by the BoVW technique. This way, we show a novel作者: 施魔法 時(shí)間: 2025-3-28 09:33
,Using BERT to?Predict the?Brazilian Stock Market,cial time series, avoiding human intervention in the process. Our results are promising, with the developed approach overcoming the baselines Buy & Hold and Moving Average Crossover, considering both profitability and risk during the investment process.作者: catagen 時(shí)間: 2025-3-28 13:22 作者: 大罵 時(shí)間: 2025-3-28 16:16
,Human Identification Based on?Gait and?Soft Biometrics,viduals by the way they walk, and this premise is the basis of most recent works on gait recognition. However, even state-of-art methods, still do not present the required robustness to work on a productive environment. The goal of this work is to propose an improvement to state-of-art gait recognit作者: nonsensical 時(shí)間: 2025-3-28 20:31
Training Aware Sigmoidal Optimization,hase uses a low learning rate to approach the center of the local minimum previously found slowly. We compared the proposed approach with commonly used adaptive learning rates schedules such as Adam, RMSProp, and Adagrad. Our experiments showed that TASO outperformed all competing methods in both op作者: 統(tǒng)治人類 時(shí)間: 2025-3-29 00:24
,A Study of?Emergency Siren Recognition on?Resource-Constrained Devices,licly available datasets: ESC-50, US8K and AudioSet. The best performing model on ESC-50 was GhostNet, achieving an F-Score of ., with a Multiply-and-Accumulate Complexity (MACC) of 0.517., whereas the 1D CNN obtained the best F-Score on US8K (.), with an MACC of 27.125.. Additionally, we verified t作者: 他去就結(jié)束 時(shí)間: 2025-3-29 03:54 作者: 大炮 時(shí)間: 2025-3-29 09:32 作者: BOGUS 時(shí)間: 2025-3-29 11:48
Event Detection in Therapy Sessions for Children with Autism,ate how machine learning techniques can help specialists by providing an automated analysis of ASD recorded therapy sessions. The proposed solution is capable of handling large amounts of video data, filtering out irrelevant frames and keeping only relevant scenes for posterior analysis. Our results作者: 考古學(xué) 時(shí)間: 2025-3-29 18:31 作者: 大都市 時(shí)間: 2025-3-29 20:01 作者: FACET 時(shí)間: 2025-3-30 00:22
Claudio Dornelles Mello Jr.,Bryan Umpierre Moreira,Paulo Jefferson Dias de Oliveira Evald,Paulo Jorg作者: 好開(kāi)玩笑 時(shí)間: 2025-3-30 04:43 作者: GUMP 時(shí)間: 2025-3-30 11:59 作者: Mawkish 時(shí)間: 2025-3-30 15:59
David Macêdo,Pedro Dreyer,Teresa Ludermir,Cleber Zanchettin作者: 身體萌芽 時(shí)間: 2025-3-30 18:30 作者: 招惹 時(shí)間: 2025-3-31 00:33
Diego Marques de Azevedo,Guilherme Souza Rodrigues,Marcelo Ladeira作者: 領(lǐng)帶 時(shí)間: 2025-3-31 01:32
Felipe Rafael de Souza,Thiago Silva Miranda,Heder Soares Bernardino作者: 分解 時(shí)間: 2025-3-31 08:56 作者: 易改變 時(shí)間: 2025-3-31 11:51
Gabriel Bianchin de Oliveira,Helio Pedrini,Zanoni Dias作者: 裝入膠囊 時(shí)間: 2025-3-31 13:53
Guilherme Ocker Ribeiro,Alexandre Soli Soares,J?nata Tyska Carvalho,Mateus Grellert作者: BLOT 時(shí)間: 2025-3-31 21:09 作者: 過(guò)多 時(shí)間: 2025-4-1 01:03 作者: enhance 時(shí)間: 2025-4-1 05:13 作者: 啟發(fā) 時(shí)間: 2025-4-1 09:39
