No abstract available.
Proceeding Downloads
A Novel Deep Learning Approach to the Statistical Downscaling of Temperatures for Monitoring Climate Change
General Circulation Models (GCMs) allow for the simulation of several climate variables through the year 2100. GCM simulations, however, are too coarse to monitor climate change at a local scale in a local region. Hence, one needs to perform spatial ...
Deep Reinforcement Learning with Noisy Exploration for Autonomous Driving
Autonomous driving decision-making is a great challenge in complex traffic environment, and the deep reinforcement learning (DRL) can contribute to the more intelligent strategy. In the autonomous driving scenarios with DRL algorithms, sufficient ...
Deep-learning based method for breech face comparisons
When a bullet is fired from a barrel, micro impression marks caused by the breech face on cartridge cases are one of the most critical factors in ballistic identification. This paper focuses on breech face impression images and introduces a deep-...
A Deep Learning-Based System for Document Layout Analysis
Document image understanding is an essential process in the digital transformation era. Those systems automatically convert a paper document to a digital document for storing and information extracting. In practice, document layout analysis is a ...
LDP: A Large Diffuse Filter Pruning to Slim the CNN
In recent years, filter pruning has become one of the most promising methods for CNN compression. However, most of the existing approaches need to prune filters iteratively, which increases the cost and complexity of the pruning process. In this paper, ...
Double Decomposition-Based Wind Speed Prediction Model for Urad Area
Renewable energy becomes progressively more important as time goes on. Wind, as one of the main rapidly developing renewable energy, is free, widely distributed, clean, environmental protection and sustainable development. The uncertainty of wind power ...
DGE-GSIM: A multi-task dual graph embedding learning for graph similarity computation
Graph similarity estimation is a challenging task due to the complex graph structure. To achieve an exact similarity estimation for input graphs, two critical factors are how to learn an appropriate graph embedding and how to compute the similarity ...
Two-Stage Dual-Archive Fireworks Algorithm for Multimodal Multi-Objective Optimization
In recent years, multi-objective optimization has attracted a lot of attention in the field of high performance computing. In this paper, two-stage dual-archive fireworks algorithm (TSDA_MMOFWA) is proposed to solve the multimodal multi-objective ...
Bi-objective lion swarm optimization based on teaching and learning algorithm
To address the problem that it is difficult to obtain a good quality and uniformly distributed Pareto optimal solution set for a complex bi-objective system model, this paper proposes a Teaching-Learning-based Bi-objective Lion Swarm Optimization ...
Music Sheet Understanding and Tone Transposition
- Khoa Minh Truong,
- Minh Cong Dinh,
- Triet Minh Huynh,
- Duc Tuan Nguyen,
- Phuc Hong Nguyen,
- Khoa Tho Anh Nguyen
Optical Music Recognition (OMR) is a sub-field in Artificial Intelligence. Automation of the translation, or understanding music sheets are the main goals of OMR. The application of this field includes the documentation of music sheets for storage or ...
Popularity Debiased Entity Linking by Adversarial Attack
Entity linking is critical for many Natural Language Processing (NLP) tasks, which aims to map textual mentions to the corresponding entities in KBs. Existing approaches have achieved promising results, however, these approaches are limited by the ...
A real time video object tracking method for fish
Fish behavior is an important indicator of water quality in smart aquaculture. The change of fish behavior can timely and effectively reflect the change of water quality. However, for the fish behavior tracking, existing tracking methods still face ...
Attentive Manifold Mixup for Model Robustness
The robustness of deep neural networks becomes more and more significant since the performance of models degrades heavily in real life. The main reason behind that is discrepancy between training and testing distribution. Many state-of-art methods have ...
Abnormal Behavior Recognition of Underwater Fish Body Based on C3D Model
The behavior of fish is the direct embodiment of fish life. It is of great significance for the management of mariculture to recognize the abnormal behavior of fish. Traditional abnormal behavior monitoring uses manual monitoring, which will cost a lot ...
Personnel status detection model suitable for vertical federated learning structure
With the improvement of the medical system, the universal access of wearable devices and people's greater concern about personal health, personnel health detection has received greater attention. However, existing personnel status detection faces the ...
Action Recognition Based on Person-Object Relationship Spatio-Temporal Graph
Human action recognition has a wide range of applications in real life. Aiming at the problem that the existing action recognition framework cannot describe the current object state of the behavior and the interaction between the object,this paper ...
Predicting Depression Symptoms from Discord Chat Messaging Using AI Medical Chatbots
Depression is a chronic illness with even Olympic athletes [1] and top tennis players [2] withdrawing from competitions due to it. It's important to diagnose depression early. Traditional methods rely on questionnaires to evaluate depression. But they ...
Transformer-Convolution Network for Arbitrary Shape Text Detection
Arbitrary shape text detection is a prevalent topic in computer vision. Text instances in natural scenes may involve different sizes, different shapes, and complex background textures. Therefore, the ability to extract accurate text features becomes ...
Recommendation Based on Graph Neural Network with Structural Identity
With the development of graph neural networks (GNN), some researchers use interaction records to construct graphs and use GNN to model and capture the information on the neighborhood of user nodes or item nodes, so as to make good use of cross-user ...
Privacy-Preserving Vertical Federated Logistic Regression without Trusted Third-Party Coordinator
Federated learning is a new distributed learning paradigm, which allows multiple parties to cooperatively train a centralized model without sharing their data. In this paper, a privacy-preserving logistic regression (LR) training algorithm for vertical ...
Knowledge Graph Entity Typing with Contrastive Learning
Knowledge graph entity typing is an important way to complete knowledge graphs (KGs), aims at predicting the associating types of certain given entities. However, previous methods suppose that many (entity, entity type) pairs can be obtained for each ...
Online Multiple Object Tracking using Physical Location Prediction
Tracking-by-detection is a commonly used paradigm for multiple-object tracking. This paper presents a method that incorporates the prediction of physical locations of people into the tracking-by-detection paradigm. The proposed method predicts the ...
A Hybrid Scheme of Reducing Read Latency in NAND Flash by Integrating Duplications and Soft Decoding
The continuous development of the technology scale has made the storage density of NAND flash memory larger and larger, and its reliability has become lower and lower. Therefore, reliability has become an issue of concern to everyone. In order to ...
Graph Convolution Word Embedding and Attention for Text Classification∗
Text classification is an important and classic task of natural language processing. Deep neural networks are becoming more and more popular in text classification due to their expressive power and low requirements for feature engineering. However, the ...
Contrastive Learning for Event Extraction
Event extraction is an important information extraction task, aiming at extracting event information from text. Each event consists of triggers and arguments with specific roles. Event extraction methods first identify the trigger and classify it into ...
Progressive Multimodal Shape Generation via Contextual Part Reasoning
We present a contextual generative network for 3D shapes based on a conditional variational autoencoder, which learns a subspace of plausible complementary parts in the context of a partial shape. With the learned part subspace prior, which encodes bi-...
Insight Evaluation on Traditional and CNN Features
Feature extraction serves as the prerequisite for any intelligent-based applications. There are various methods to extract features from the initial data sets, namely traditional filters and deep learning components. However, current traditional ...
Index Terms
- Proceedings of the 2022 6th International Conference on Machine Learning and Soft Computing