scholarly journals Multitask Learning over Shared Subspaces

2020 ◽  
Author(s):  
Nicholas Menghi ◽  
Kemal Kacar ◽  
Will Penny
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Transfer Learning ◽  
Network Model ◽  
Neural Network Model ◽  
Human Performance ◽  
Human Subjects ◽  
Model Learning ◽  
Learning Tasks ◽  
Parameter Sharing

AbstractThis paper uses constructs from the field of multitask machine learning to define pairs of learning tasks that either shared or did not share a common subspace. Human subjects then learnt these tasks using a feedback-based approach. We found, as hypothesised, that subject performance was significantly higher on the second task if it shared the same subspace as the first, an advantage that played out most strongly at the beginning of the second task. Additionally, accuracy was positively correlated over subjects learning same-subspace tasks but was not correlated for those learning different-subspace tasks. These results, and other aspects of learning dynamics, were compared to the behaviour of a Neural Network model trained using sequential Bayesian inference. Human performance was found to be consistent with a Soft Parameter Sharing variant of this model that constrained representations to be similar among tasks but only when this aided learning. We propose that the concept of shared subspaces provides a useful framework for the experimental study of human multitask and transfer learning.Author summaryHow does knowledge gained from previous experience affect learning of new tasks ? This question of “Transfer Learning” has been addressed by teachers, psychologists, and more recently by researchers in the fields of neural networks and machine learning. Leveraging constructs from machine learning, we designed pairs of learning tasks that either shared or did not share a common subspace. We compared the dynamics of transfer learning in humans with those of a multitask neural network model, finding that human performance was consistent with a soft parameter sharing variant of the model. Learning was boosted in the early stages of the second task if the same subspace was shared between tasks. Additionally, accuracy between tasks was positively correlated but only when they shared the same subspace. Our results highlight the roles of subspaces, showing how they could act as a learning boost if shared, and be detrimental if not.

scholarly journals PREDICTION AND MAPPING OF COCHLODINIUM POLYKRIKOIDES RED TIDE USING MACHINE LEARNING UNDER IMBALANCED DATA

2019 ◽  
Vol XLII-4/W16 ◽  
pp. 143-145
Author(s):  
S. H. Bak ◽  
D. H. Hwang ◽  
U. Enkhjargal ◽  
H. J. Yoon
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Network Model ◽  
Neural Network Model ◽  
Red Tide ◽  
Imbalanced Data ◽  
Cochlodinium Polykrikoides ◽  
Model Learning ◽  
Data Set ◽  
Data Problem

Abstract. Cochlodinium polykrikoides (C. polykrikoides) is a phytoplankton that causes red tides every year in the middle of the South Sea of Korea. C. polykrikoides is a harmful Algae that has migratory ability and causes the fisheries damage over a long period of wide sea area if it causes red tide once. To minimize red tide damage, it is important to anticipate and prepare the red tide occurrence timing and location in advance. In this study, we predicted the occurrence of red tide of C. polykrikoides using machine learning techniques and compared the results of each algorithm. Logistic regression model, decision tree model, and multilayer neural network model were used for prediction of red tide occurrence. To produce the data set for model learning, we used the red tide occurrence map provided by the National Institute of Fisheries Science, the Local Data Assimilation and Prediction System (LDAPS) provided by the Korea Meteorological Agency, and the G1SST provided by the National Oceanic and Atmospheric Administration (NOAA). The feature vectors used for modeling consisted of 59 elements, which were made by using temperature, water temperature, precipitation, solar radiation, wind direction and wind speed. Only a very small number of red tide cases can be collected compared to the case of no red tide cases. Thus, an imbalance data problem arises in the data set. To overcome this imbalanced data problem, we used adding noise after oversampling to data of red tide occurrence to solve the difference of data between two classes.The data set is divided into 8 : 2 to prevent over-fitting and 80% is used as the learning data. The remaining 20% was used to evaluate the performance of each model. As a result of evaluating the prediction performance of each model, the multilayer neural network model showed the highest prediction accuracy.

scholarly journals Burst Pressure Prediction of API 5L X-Grade Dented Pipelines Using Deep Neural Network

2020 ◽  
Vol 8 (10) ◽  
pp. 766
Author(s):  
Dohan Oh ◽  
Julia Race ◽  
Selda Oterkus ◽  
Bonguk Koo
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Artificial Neural Network ◽  
Network Model ◽  
Neural Network Model ◽  
Deep Neural Network ◽  
Machine Learning Techniques ◽  
Burst Pressure ◽  
Comparison Results ◽  
Artificial Neural

Mechanical damage is recognized as a problem that reduces the performance of oil and gas pipelines and has been the subject of continuous research. The artificial neural network in the spotlight recently is expected to be another solution to solve the problems relating to the pipelines. The deep neural network, which is on the basis of artificial neural network algorithm and is a method amongst various machine learning methods, is applied in this study. The applicability of machine learning techniques such as deep neural network for the prediction of burst pressure has been investigated for dented API 5L X-grade pipelines. To this end, supervised learning is employed, and the deep neural network model has four layers with three hidden layers, and the neural network uses the fully connected layer. The burst pressure computed by deep neural network model has been compared with the results of finite element analysis based parametric study, and the burst pressure calculated by the experimental results. According to the comparison results, it showed good agreement. Therefore, it is concluded that deep neural networks can be another solution for predicting the burst pressure of API 5L X-grade dented pipelines.

scholarly journals Nowcasting of Lumber Futures Price with Google Trends Index Using Machine Learning and Deep Learning Models

2021 ◽  
Vol 72 (1) ◽  
pp. 11-20
Author(s):  
Mingtao He ◽  
Wenying Li ◽  
Brian K. Via ◽  
Yaoqi Zhang
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Deep Learning ◽  
Network Model ◽  
Neural Network Model ◽  
Price Volatility ◽  
Google Trends ◽  
Learning Models ◽  
Fresh Perspective ◽  
Futures Price

Abstract Firms engaged in producing, processing, marketing, or using lumber and lumber products always invest in futures markets to reduce the risk of lumber price volatility. The accurate prediction of real-time prices can help companies and investors hedge risks and make correct market decisions. This paper explores whether Internet browsing habits can accurately nowcast the lumber futures price. The predictors are Google Trends index data related to lumber prices. This study offers a fresh perspective on nowcasting the lumber price accurately. The novel outlook of employing both machine learning and deep learning methods shows that despite the high predictive power of both the methods, on average, deep learning models can better capture trends and provide more accurate predictions than machine learning models. The artificial neural network model is the most competitive, followed by the recurrent neural network model.

scholarly journals Image Classification Using Transfer Learning and Deep Learning

2021 ◽  
Vol 10 (9) ◽  
pp. 25394-25398
Author(s):  
Chitra Desai
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Image Classification ◽  
Transfer Learning ◽  
Network Model ◽  
Neural Network Model ◽  
Large Scale ◽  
Deep Neural Network ◽  
Visual Recognition ◽  
Classification Of Images

Deep learning models have demonstrated improved efficacy in image classification since the ImageNet Large Scale Visual Recognition Challenge started since 2010. Classification of images has further augmented in the field of computer vision with the dawn of transfer learning. To train a model on huge dataset demands huge computational resources and add a lot of cost to learning. Transfer learning allows to reduce on cost of learning and also help avoid reinventing the wheel. There are several pretrained models like VGG16, VGG19, ResNet50, Inceptionv3, EfficientNet etc which are widely used.   This paper demonstrates image classification using pretrained deep neural network model VGG16 which is trained on images from ImageNet dataset. After obtaining the convolutional base model, a new deep neural network model is built on top of it for image classification based on fully connected network. This classifier will use features extracted from the convolutional base model.

scholarly journals Cloud based mobile solution for early detection of Skin Cancer using Artificial Intelligence

2021 ◽  
pp. 312-324
Author(s):  
Pawan Sonawane ◽  
Sahel Shardhul ◽  
Raju Mendhe
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Skin Cancer ◽  
Early Detection ◽  
Transfer Learning ◽  
Network Model ◽  
Neural Network Model ◽  
Mobile Application ◽  
Low Cost ◽  
Skin Tumors

The vast majority of skin cancer deaths are from melanoma, with about 1.04 million cases annually. Early detection of the same can be immensely helpful in order to try to cure it. But most of the diagnosis procedures are either extremely expensive or not available to a vast majority, as these centers are concentrated in urban regions only. Thus, there is a need for an application that can perform a quick, efficient, and low-cost diagnosis. Our solution proposes to build a server less mobile application on the AWS cloud that takes the images of potential skin tumors and classifies it as either Malignant or Benign. The classification would be carried out using a trained Convolution Neural Network model and Transfer learning (Inception v3). Several experiments will be performed based on Morphology and Color of the tumor to identify ideal parameters.

scholarly journals Real-Time Optimization and Control of Nonlinear Processes Using Machine Learning

Mathematics ◽  
2019 ◽  
Vol 7 (10) ◽  
pp. 890 ◽  
Author(s):  
Zhihao Zhang ◽  
Zhe Wu ◽  
David Rincon ◽  
Panagiotis Christofides
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Real Time ◽  
Network Model ◽  
Neural Network Model ◽  
First Principles ◽  
Chemical Process ◽  
Feed Forward Neural Network ◽  
Time Optimization ◽  
Real Time Optimization

Machine learning has attracted extensive interest in the process engineering field, due to the capability of modeling complex nonlinear process behavior. This work presents a method for combining neural network models with first-principles models in real-time optimization (RTO) and model predictive control (MPC) and demonstrates the application to two chemical process examples. First, the proposed methodology that integrates a neural network model and a first-principles model in the optimization problems of RTO and MPC is discussed. Then, two chemical process examples are presented. In the first example, a continuous stirred tank reactor (CSTR) with a reversible exothermic reaction is studied. A feed-forward neural network model is used to approximate the nonlinear reaction rate and is combined with a first-principles model in RTO and MPC. An RTO is designed to find the optimal reactor operating condition balancing energy cost and reactant conversion, and an MPC is designed to drive the process to the optimal operating condition. A variation in energy price is introduced to demonstrate that the developed RTO scheme is able to minimize operation cost and yields a closed-loop performance that is very close to the one attained by RTO/MPC using the first-principles model. In the second example, a distillation column is used to demonstrate an industrial application of the use of machine learning to model nonlinearities in RTO. A feed-forward neural network is first built to obtain the phase equilibrium properties and then combined with a first-principles model in RTO, which is designed to maximize the operation profit and calculate optimal set-points for the controllers. A variation in feed concentration is introduced to demonstrate that the developed RTO scheme can increase operation profit for all considered conditions.

Sign in / Sign up

Export Citation Format

Share Document