A Survey of Deep Learning: Platforms, Applications and Emerging Research Trends

Hatcher, William G.; Yu, Wei

doi:10.1109/access.2018.2830661

Cited by 531 publications

(210 citation statements)

References 125 publications

Supporting

Mentioning

209

Contrasting

Unclassified

Order By: Relevance

“…For that reason, "properly controlling or regularizing the training is key to outof-sample generalization" (Zhang et al 2018). • Deep learning As Hatcher and Yu point out, DL applies "multi-neuron, multi-layer neural networks to perform learning tasks, including regression, classification, clustering, auto-encoding, and others" (Hatcher and Yu 2018). As a specific form of representation learning, which in turn is a specific form of ML, DL is based on artificial neural networks (ANNs) such as convolutional neural networks (CNNs) ).…”

Section: Figmentioning

confidence: 99%

Geospatial Artificial Intelligence: Potentials of Machine Learning for 3D Point Clouds and Geospatial Digital Twins

Döllner

2020

PFG

View full text Add to dashboard Cite

Artificial intelligence (AI) is changing fundamentally the way how IT solutions are implemented and operated across all application domains, including the geospatial domain. This contribution outlines AI-based techniques for 3D point clouds and geospatial digital twins as generic components of geospatial AI. First, we briefly reflect on the term "AI" and outline technology developments needed to apply AI to IT solutions, seen from a software engineering perspective. Next, we characterize 3D point clouds as key category of geodata and their role for creating the basis for geospatial digital twins; we explain the feasibility of machine learning (ML) and deep learning (DL) approaches for 3D point clouds. In particular, we argue that 3D point clouds can be seen as a corpus with similar properties as natural language corpora and formulate a "Naturalness Hypothesis" for 3D point clouds. In the main part, we introduce a workflow for interpreting 3D point clouds based on ML/ DL approaches that derive domain-specific and application-specific semantics for 3D point clouds without having to create explicit spatial 3D models or explicit rule sets. Finally, examples are shown how ML/DL enables us to efficiently build and maintain base data for geospatial digital twins such as virtual 3D city models, indoor models, or building information models.

show abstract

Section: Figmentioning

confidence: 99%

Geospatial Artificial Intelligence: Potentials of Machine Learning for 3D Point Clouds and Geospatial Digital Twins

Döllner

2020

PFG

View full text Add to dashboard Cite

show abstract

“…For each state experienced by the agent, it has to remember it and perform an experience replay. The memory contains tuples of state, next state, action, reward and a Boolean value for indicating the termination of the agent [3]. This keeps on going and the agent memorizes the information until the termination happens as in Fig.2.…”

Section: Deep Q Learningmentioning

confidence: 99%

Deep Q Learning in Stabilization of Inverted Pendulum

Amudhan*,

Sedani,

Dwivedi

2019

IJITEE

View full text Add to dashboard Cite

Reinforcement Learning is a concept in which an agent takes actions in an environment to maximize its aggregate reward. Agents in Reinforcement Learning algorithms are penalized for bad actions and rewarded for good ones. Deep Reinforcement learning is used to overcome some shortcomings in Reinforcement learning like lack of variety in the open source collection and change in task’s difficulty based on reward or set of actions. In a larger environment, inference of new states from already explored states is a difficult task due to its time & space complexity. Hence the Q-value function which represents the quality value is approximated by Deep Q-learning that uses a neural network for the same. The results of implementing Deep Q Learning (DQN), Double DQN, Dueling DQN, Noisy DQN & DQN Prioritized Experience Replay techniques & their performance in stabilization of inverted pendulum are highlighted in this paper. Deep Reinforcement learning can be applied in various platform like for recognition, perception in computer vision, for simulation to real robot control in robotics, for sequence generation, translation in Natural Language processing, for Poker, Bridge StarCraft in games, for pricing, trading risk management in finance, for e- commerce, customer management in business management, for diagnosis using Electronic Medical Records in healthcare and in adaptive decision control.

show abstract

“…There have been accounted for certain situations when the IoT gadgets were not possibly solid to shield their code and information from outside access which in the long run makes the assailant to clone whole gadget or control the product or information. Maybe a couple of the models are the physical security assault when several brilliant traffic light gadgets were harmed by hoodlums who took the SIM cards of gadgets [17,18]. Those SIM cards were later used to make cell phone brings in South Africa alongside a few vehicle crashes at the area and an extra cost to fix the whole framework.…”

Section: Other Threats and Issuesmentioning

confidence: 99%

Secured Authentication Systems for Internet of Things

Gowtham¹,

Banga²,

Patil³

2018

EAI Endorsed Transactions on Smart Cities

View full text Add to dashboard Cite

INTRODUCTION: In these days, an enormous extent of contraptions are interconnected with the remote advances which gave the principal light to the front line development of Internet of Things (IoT). Different quick contraptions and machines are by and by watched and controlled using IoT conventions. The developments of IoT are by and by spread to the entire sphere by which there is superb system in the devices related using IoT. From the assessment reports of Statista.com, The closeout of splendid home contraptions raised from 1.2 billion dollars to 4.4 billion dollars from year 2015 to year 2019 in the United States. As indicated by the report from Economics Times, there will connect with more than 2 billion units of ESIM based contraptions by year 2024. With the use of ESIM, the endorsers can use the progressed SIM card for the astute contraptions and the organizations can be activated without need of the physical SIM card. It is one of the progressing and confirmed employments of Internet of Things (IoT). OBJECTIVES: The presented research manuscript is presenting an outline of the present state of IoT security. METHODS: Past the standard applications, IoT is under research for the earth watching and prior notification to the coordinating workplaces so the fitting moves can be made. As per the news report by Grand View Research Inc., the overall IoT marketplace size is shown to contact more than 5,000 million dollars by year 2025. The presented IoT suggests the radio advancement standard with LPWAN so the enormous consideration of sharp devices should be conceivable with more significant level of execution in the system. RESULTS: The key positive of the paper integrates the evaluation of Internet of Things with the assorted dimensions in addition to the cavernous analytics with the implementation aspects towards the security mechanism. The paper is having the focus and goals towards the association of security aware mechanism for the cumulative performance of IoT based environment. CONCLUSION: With the gigantic utilization of IoT, there is have to incorporate the higher level of security and honesty for the protection mindful system condition.

show abstract

A Survey of Deep Learning: Platforms, Applications and Emerging Research Trends

Cited by 531 publications

References 125 publications

Geospatial Artificial Intelligence: Potentials of Machine Learning for 3D Point Clouds and Geospatial Digital Twins

Geospatial Artificial Intelligence: Potentials of Machine Learning for 3D Point Clouds and Geospatial Digital Twins

Deep Q Learning in Stabilization of Inverted Pendulum

Secured Authentication Systems for Internet of Things

Contact Info

Product

Resources

About