Deep learning has exploded in the public consciousness, primarily as predictive and analytical products suffuse our world, in the form of numerous human-centered smart-world systems, including targeted advertisements, natural language assistants and interpreters, and prototype self-driving vehicle systems. Yet to most, the underlying mechanisms that enable such human-centered smart products remain obscure. In contrast, researchers across disciplines have been incorporating deep learning into their research to solve problems that could not have been approached before. In this paper, we seek to provide a thorough investigation of deep learning in its applications and mechanisms. Specifically, as a categorical collection of state of the art in deep learning research, we hope to provide a broad reference for those seeking a primer on deep learning and its various implementations, platforms, algorithms, and uses in a variety of smart-world systems. Furthermore, we hope to outline recent key advancements in the technology, and provide insight into areas, in which deep learning can improve investigation, as well as highlight new areas of research that have yet to see the application of deep learning, but could nonetheless benefit immensely. We hope this survey provides a valuable reference for new deep learning practitioners, as well as those seeking to innovate in the application of deep learning.

https://ieeexplore.ieee.org/ielaam/6287639/8274985/8351898-aam.pdf

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

Deep learning models for traffic flow prediction in autonomous vehicles: A review, solutions, and challenges

Deep reinforcement learning with smooth policy update: Application to robotic cloth manipulation

In this letter, we explore using self-supervised correspondence for improving the generalization performance and sample efficiency of visuomotor policy learning. Prior work has primarily used approaches such as autoencoding, pose-based losses, and end-to-end policy optimization in order to train the visual portion of visuomotor policies. We instead propose an approach using self-supervised dense visual correspondence training and show that this enables visuomotor policy learning with surprisingly high generalization performance with modest amounts of data. Using imitation learning, we demonstrate extensive hardware validation on challenging manipulation tasks with as few as 50 demonstrations. Our learned policies can generalize across classes of objects, react to deformable object configurations, and manipulate textureless symmetrical objects in a variety of backgrounds, all with closed-loop, real-time vision-based policies. Simulated imitation learning experiments suggest that correspondence training offers sample complexity and generalization benefits compared to autoencoding and end-to-end training.

Self-Supervised Correspondence in Visuomotor Policy Learning

This study proposes the use of a deep neural network to localize a sound source using an array of microphones in a reverberant environment. During the last few years, applications based on deep neural networks have performed various tasks such as image classification or speech recognition to levels that exceed even human capabilities. In our study, we employ deep residual networks, which have recently shown remarkable performance in image classification tasks even when the training period is shorter than that of other models. Deep residual networks are used to process audio input similar to multiple signal classification (MUSIC) methods. We show that with end-to-end training and generic preprocessing, the performance of deep residual networks not only surpasses the block level accuracy of linear models on nearly clean environments but also shows robustness to challenging conditions by exploiting the time delay on power information.

https://www.jstage.jst.go.jp/article/jrobomech/29/1/29_37/_pdf

Sound source localization using deep learning models

We propose a practical state-of-the-art method to develop a machine-learning-based humanoid robot that can work as a production line worker. The proposed approach provides an intuitive way to collect data and exhibits the following characteristics: task performing capability, task reiteration ability, generalizability, and easy applicability. The proposed approach utilizes a real-time user interface with a monitor and provides a first-person perspective using a head-mounted display. Through this interface, teleoperation is used for collecting task operating data, especially for tasks that are difficult to be applied with a conventional method. A two-phase deep learning model is also utilized in the proposed approach. A deep convolutional autoencoder extracts images features and reconstructs images, and a fully connected deep time delay neural network learns the dynamics of a robot task process from the extracted image features and motion angle signals. The “Nextage Open” humanoid robot is used as an experimental platform to evaluate the proposed model. The object folding task utilizing with 35 trained and 5 untrained sensory motor sequences for test. Testing the trained model with online generation demonstrates a 77.8% success rate for the object folding task.

Repeatable Folding Task by Humanoid Robot Worker Using Deep Learning

While robots can learn models to solve many manipulation tasks from raw visual input, they cannot usually use these models to solve new problems. On the other hand, symbolic planning methods such as STRIPS have long been able to solve new problems given only a domain definition and a symbolic goal, but these approaches often struggle on the real world robotic tasks due to the challenges of grounding these symbols from sensor data in a partially-observable world. We propose Deep Planning Domain Learning (DPDL), an approach that combines the strengths of both methods to learn a hierarchical model. DPDL learns a high-level model which predicts values for a large set of logical predicates consisting of the current symbolic world state, and separately learns a low-level policy which translates symbolic operators into executable actions on the robot. This allows us to perform complex, multistep tasks even when the robot has not been explicitly trained on them. We show our method on manipulation tasks in a photorealistic kitchen scenario.

Transferable Task Execution from Pixels through Deep Planning Domain Learning

For robots to have a wide range of applications, they must be able to execute numerous tasks. However, recent studies into robot manipulation using deep neural networks (DNN) have primarily focused on single tasks. Therefore, we investigate a robot manipulation model that uses DNNs and can execute long sequential dynamic tasks by performing multiple short sequential tasks at appropriate times. To generate compound tasks, we propose a model comprising two DNNs: a convolutional autoencoder that extracts image features and a multiple timescale recurrent neural network (MTRNN) to generate motion. The internal state of the MTRNN is constrained to have similar values at the initial and final motion steps; thus, motions can be differentiated based on the initial image input. As an example compound task, we demonstrate that the robot can generate a “Put-In-Box” task that is divided into three subtasks: open the box, grasp the object and put it into the box, and close the box. The subtasks were trained as discrete tasks, and the connections between each subtask were not trained. With the proposed model, the robot could perform the Put-In-Box task by switching among subtasks and could skip or repeat subtasks depending on the situation.

Put-in-Box Task Generated from Multiple Discrete Tasks by aHumanoid Robot Using Deep Learning

While robots can learn models to solve many manipulation tasks from raw visual input, they cannot usually use these models to solve new problems. On the other hand, symbolic planning methods such as STRIPS have long been able to solve new problems given only a domain definition and a symbolic goal, but these approaches often struggle on the real world robotic tasks due to the challenges of grounding these symbols from sensor data in a partially-observable world. We propose Deep Planning Domain Learning (DPDL), an approach that combines the strengths of both methods to learn a hierarchical model. DPDL learns a high-level model which predicts values for a large set of logical predicates consisting of the current symbolic world state, and separately learns a low-level policy which translates symbolic operators into executable actions on the robot. This allows us to perform complex, multi-step tasks even when the robot has not been explicitly trained on them. We show our method on manipulation tasks in a photorealistic kitchen scenario.

Kei Kase

Papers

Repeatable Folding Task by Humanoid Robot Worker Using Deep Learning

Transferable Task Execution from Pixels through Deep Planning Domain Learning

Put-in-Box Task Generated from Multiple Discrete Tasks by aHumanoid Robot Using Deep Learning

Transferable Task Execution from Pixels through Deep Planning Domain Learning

Put-In-Box task generated from multiple discrete tasks by humanoid robot using deep learning