Adaptive Model Scheduling for Resource-efficient Data Labeling

Labeling data (e.g., labeling the people, objects, actions, and scene in images) comprehensively and efficiently is a widely needed but challenging task. Numerous models were proposed to label various data and many approaches were designed to enhance the ability of deep learning models or accelerate them. Unfortunately, a single machine-learning model is not powerful enough to extract various semantic information from data. Given certain applications, such as image retrieval platforms and photo album management apps, it is often required to execute a collection of models to obtain sufficient labels. With limited computing resources and stringent delay, given a data stream and a collection of applicable resource-hungry deep-learning models, we design a novel approach to adaptively schedule a subset of these models to execute on each data item, aiming to maximize the value of the model output (e.g., the number of high-confidence labels). Achieving this lofty goal is nontrivial since a model’s output on any data item is content-dependent and unknown until we execute it. To tackle this, we propose an Adaptive Model Scheduling framework, consisting of (1) a deep reinforcement learning-based approach to predict the value of unexecuted models by mining semantic relationship among diverse models, and (2) two heuristic algorithms to adaptively schedule the model execution order under a deadline or deadline-memory constraints, respectively. The proposed framework does not require any prior knowledge of the data, which works as a powerful complement to existing model optimization technologies. We conduct extensive evaluations on five diverse image datasets and 30 popular image labeling models to demonstrate the effectiveness of our design: our design could save around 53% execution time without loss of any valuable labels.

Can Computers Outperform Humans in Detecting User Zone-Outs? Implications for Intelligent Interfaces

The ability to identify whether a user is “zoning out” (mind wandering) from video has many HCI (e.g., distance learning, high-stakes vigilance tasks). However, it remains unknown how well humans can perform this task, how they compare to automatic computerized approaches, and how a fusion of the two might improve accuracy. We analyzed videos of users’ faces and upper bodies recorded 10s prior to self-reported mind wandering (i.e., ground truth) while they engaged in a computerized reading task. We found that a state-of-the-art machine learning model had comparable accuracy to aggregated judgments of nine untrained human observers (area under receiver operating characteristic curve [AUC] = .598 versus .589). A fusion of the two (AUC = .644) outperformed each, presumably because each focused on complementary cues. Furthermore, adding more humans beyond 3–4 observers yielded diminishing returns. We discuss implications of human–computer fusion as a means to improve accuracy in complex tasks.

Privacy Rating of Mobile Applications Based on Crowdsourcing and Machine Learning

With the advent of the 5G network era, the convenience of mobile smartphones has become increasingly prominent, the use of mobile applications has become wider and wider, and the number of mobile applications. However, the privacy of mobile applications and the security of users' privacy information are worrying. This article aims to study the ratings of data and machine learning on the privacy security of mobile applications, and uses the experiments in this article to conduct data collection, data analysis, and summary research. This paper experimentally establishes a machine learning model to realize the prediction of privacy scores of Android applications. The establishment of this model is based on the intent of using sensitive permissions in the application and related metadata. It is to create a regression function that can implement the mapping of applications to score . Experimental data shows that the feature vector prediction model can uniquely be used to represent the actual usage and scheme of a system's specific permissions for the application.

Privacy Rating of Mobile Applications Based on Crowdsourcing and Machine Learning

With the advent of the 5G network era, the convenience of mobile smartphones has become increasingly prominent, the use of mobile applications has become wider and wider, and the number of mobile applications. However, the privacy of mobile applications and the security of users' privacy information are worrying. This article aims to study the ratings of data and machine learning on the privacy security of mobile applications, and uses the experiments in this article to conduct data collection, data analysis, and summary research. This paper experimentally establishes a machine learning model to realize the prediction of privacy scores of Android applications. The establishment of this model is based on the intent of using sensitive permissions in the application and related metadata. It is to create a regression function that can implement the mapping of applications to score . Experimental data shows that the feature vector prediction model can uniquely be used to represent the actual usage and scheme of a system's specific permissions for the application.

Prediction of Credit Card Approval

Credit risk as the board in banks basically centers around deciding the probability of a customer's default or credit decay and how expensive it will end up being assuming it happens. It is important to consider major factors and predict beforehand the probability of consumers defaulting given their conditions. Which is where a machine learning model comes in handy and allows the banks and major financial institutions to predict whether the customer, they are giving the loan to, will default or not. This project builds a machine learning model with the best accuracy possible using python. First we load and view the dataset. The dataset has a combination of both mathematical and non-mathematical elements, that it contains values from various reaches, in addition to that it contains a few missing passages. We preprocess the dataset to guarantee the AI model we pick can make great expectations. After the information is looking great, some exploratory information examination is done to assemble our instincts. Finally, we will build a machine learning model that can predict if an individual's application for a credit card will be accepted. Using various tools and techniques we then try to improve the accuracy of the model. This project uses Jupyter notebook for python programming to build the machine learning model. Using Data Analysis and Machine Learning, we attempted to determine the most essential parameters for obtaining credit card acceptance in this project. The machine learning model we built gave an 86 % accuracy for predicting whether the credit card will be approved or not, considering the various factors mentioned in the application of the credit card holder. Even though we achieved an accuracy of 86%, we conducted a grid search to see if we could increase the performance even further. However, using both the machine learning models: random forest and logistic regression, the best we could get from this data was 86 percent.