A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Filters
Goal-constrained Sparse Reinforcement Learning for End-to-End Driving
[article]
2021
arXiv
pre-print
Preserved Fulltext
Other Versions
Deep reinforcement Learning for end-to-end driving is limited by the need of complex reward engineering. ...
In this work, we explore full-control driving with only goal-constrained sparse reward and propose a curriculum learning approach for end-to-end driving using only navigation view maps that benefit from ...
Overview of our reinforcement learning framework for end to end driving with sparse reward. ...
arXiv:2103.09189v2
fatcat:tbodnetl6beltbpmsxtqtq6viq
Accelerating Reinforcement Learning for Reaching using Continuous Curriculum Learning
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
To this end, we explore various continuous curriculum strategies for controlling a training process. ...
Reinforcement learning has shown great promise in the training of robot behavior due to the sequential decision making characteristics. ...
ACKNOWLEDGMENT We are thankful to Dr. Marco Wiering for his thoughtful suggestions. ...
arXiv:2002.02697v1
fatcat:2hlk46k5mzdunnylbsxe23kiii
Agent Coordination in Air Combat Simulation using Multi-Agent Deep Reinforcement Learning
2020
2020 IEEE International Conference on Systems, Man, and Cybernetics (SMC)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
In this work, we study how multi-agent deep reinforcement learning can be used to construct behavior models for synthetic pilots in air combat simulation. ...
We empirically evaluate a number of approaches in two air combat scenarios, and demonstrate that curriculum learning is a promising approach for handling the high-dimensional state space of the air combat ...
To help the agents learn in spite of the sparse rewards and large search area, we use curriculum learning. ...
doi:10.1109/smc42975.2020.9283492
fatcat:lqvskjcw6zhnholgmqjd7oqp7q
Deep Reinforcement Learning for Complex Manipulation Tasks with Sparse Feedback
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Learning optimal policies from sparse feedback is a known challenge in reinforcement learning. ...
Hindsight Experience Replay (HER) is a multi-goal reinforcement learning algorithm that comes to solve such tasks. ...
Reinforcement Learning for Sparse Reward Function Reward functions can be divided into two categories: Dense rewards and Sparse rewards (also known as binary rewards). ...
arXiv:2001.03877v1
fatcat:7ucagopk6vbxjlssxnqimfllum
Autonomous Overtaking in Gran Turismo Sport Using Curriculum Reinforcement Learning
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Other Versions
By leveraging curriculum learning, our approach leads to faster convergence as well as increased performance compared to vanilla reinforcement learning. ...
driving. ...
The key is to leverage taskspecific curriculum RL and a novel reward formulation to train an end-to-end neural network controller. ...
arXiv:2103.14666v2
fatcat:onysxa2af5e5bawgmah5cd3ziy
Competitive Experience Replay
[article]
2019
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Our method asymmetrically augments these sparse rewards for a pair of agents each learning the same task, creating a competitive game designed to drive exploration. ...
Deep learning has achieved remarkable successes in solving challenging reinforcement learning (RL) problems when dense reward function is provided. ...
SPARSE REWARD REINFORCEMENT LEARNING Reinforcement learning considers the problem of finding an optimal policy for an agent that interacts with an uncertain environment and collects reward per action. ...
arXiv:1902.00528v4
fatcat:pxqopbrljzeltabolip6ticriy
Accuracy-based Curriculum Learning in Deep Reinforcement Learning
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
sampled randomly learns more efficiently than when asked to be very accurate at all times. ...
In this paper, we investigate a new form of automated curriculum learning based on adaptive selection of accuracy requirements, called accuracy-based curriculum learning. ...
requirement in the end, for which it is still making progress. ...
arXiv:1806.09614v2
fatcat:wq2ukixqujbwndqhdncw3nlwxq
Fully Autonomous Real-World Reinforcement Learning with Applications to Mobile Manipulation
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Our aim is to devise a robotic reinforcement learning system for learning navigation and manipulation together, in an autonomous way without human intervention, enabling continual learning under realistic ...
While reinforcement learning in principle provides for automated robotic skill learning, in practice reinforcement learning in the real world is challenging and often requires extensive instrumentation ...
end if
31: end if
32: if rg = 1 then:
33: Drop object randomly in grasping area.
34: end if
35: end for
36: return rmax
policy, with a sparse reward, and uses the same action space. ...
arXiv:2107.13545v3
fatcat:jwenctwrb5fhllvwreqbalhf7a
Learning of Long-Horizon Sparse-Reward Robotic Manipulator Tasks with Base Controllers
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Deep Reinforcement Learning (DRL) enables robots to perform some intelligent tasks end-to-end. However, there are still many challenges for long-horizon sparse-reward robotic manipulator tasks. ...
In this paper, we propose a method of learning long-horizon sparse-reward tasks utilizing one or more existing traditional controllers named base controllers in this paper. ...
“End to end learning for self-driving cars,” NIPS 2016 Deep Learning
[5] H. Li, Q. Zhang, and D. Zhao, “Deep reinforcement learning-based Symposium, 2016. ...
arXiv:2011.12105v3
fatcat:qgjdebnjmvgadkzdkn4vgqbpsm
Automatic Curriculum Learning For Deep RL: A Short Survey
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Automatic Curriculum Learning (ACL) has become a cornerstone of recent successes in Deep Reinforcement Learning (DRL).These methods shape the learning trajectories of agents by challenging them with tasks ...
The ambition of this work is dual: 1) to present a compact and accessible introduction to the Automatic Curriculum Learning literature and 2) to draw a bigger picture of the current state of the art in ...
Deep Reinforcement Learning is a family of algorithms which leverage deep neural networks for function approximation to tackle reinforcement learning problems. ...
arXiv:2003.04664v2
fatcat:lhire3htmnenfetx2ry4furgyy
Reinforcement Learning with Iterative Reasoning for Merging in Dense Traffic
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
We design a training curriculum for a reinforcement learning agent using the concept of level-k behavior. ...
In this work, we propose a combination of reinforcement learning and game theory to learn merging behaviors. ...
CONCLUSIONS This paper presented a reinforcement learning curriculum based on level-k reasoning to learn to merge in dense traffic. ...
arXiv:2005.11895v1
fatcat:q7f74l4pjjgnxmhkl3tsh42upm
Expert-augmented actor-critic for ViZDoom and Montezumas Revenge
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
In a number of experiments, we have observed an unreported bug in Montezumas Revenge which allowed the agent to score more than 800,000 points. ...
We propose an expert-augmented actor-critic algorithm, which we evaluate on two environments with sparse rewards: Montezumas Revenge and a demanding maze from the ViZDoom suite. ...
be seen as a curriculum learning. ...
arXiv:1809.03447v1
fatcat:pn2vsmnkwrg6feq2wjjupq5nhi
Independent Generative Adversarial Self-Imitation Learning in Cooperative Multiagent Systems
[article]
2019
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Besides, we put forward a Sub-Curriculum Experience Replay mechanism to pick out the past beneficial experiences as much as possible and accelerate the self-imitation learning process. ...
Many tasks in practice require the collaboration of multiple agents through reinforcement learning. ...
is similar to the idea of reverse curriculum generation for reinforcement learning [10] . ...
arXiv:1909.11468v1
fatcat:glwziahczfcnbfkzgw4itviuli
Automating Vehicles by Deep Reinforcement Learning using Task Separation with Hill Climbing
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Within the context of autonomous driving a model-based reinforcement learning algorithm is proposed for the design of neural network-parameterized controllers. ...
The contribution of this paper is the proposition of a simple gradient-free and model-based algorithm for deep reinforcement learning using task separation with hill climbing (TSHC). ...
This motivates to leverage available vehicle models for control design. Consider also the position paper [22] for general limitations of end-to-end learning. ...
arXiv:1711.10785v2
fatcat:ymphnpudpneirff5ewgulypzma
Accelerating Training in Pommerman with Imitation and Reinforcement Learning
[article]
2019
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
The basic PPO approach is modified for stable transition from the imitation learning phase through reward shaping, action filters based on heuristics, and curriculum learning. ...
The proposed methodology is able to beat heuristic and pure reinforcement learning baselines with a combined 100,000 training games, significantly faster than other non-tree-search methods in literature ...
Introduction Reinforcement learning has achieved success in solving several complex problems, ranging from game playing [1, 2] to robotics [3] and autonomous driving [4] . ...
arXiv:1911.04947v2
fatcat:k73v3b6chfccpjbkw7ncl6mgei
« Previous
Showing results 1 — 15 out of 5,209 results