Reinforcement Learning for Robots Using Neural Networks

By Longxin Lin. 1992

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Reinforcement Learning

Citations in the corpus

Playing Atari with Deep Reinforcement Learning (2013, methodology): “l inputs; as a result it may learn features that are directly relevant to discriminating action-values. Q-learning has also previously been combined with experience replay and a simple neural network [13], but again starting with a low-dimensional state rather than raw visual inputs. The use of the ”
PGQ: Combining policy gradient and Q-learning (2016, background): “Reinforcement learning has seen success in several areas including robotics (Lin, 1993; Levine et al., 2015), computer games (Mnih et al.”
Off-Policy Actor-Critic with Shared Experience Replay (2019, background): “Experience replay (Lin, 1992) mixes data from multiple iterations of policy improvement.”
Recurrent World Models Facilitate Policy Evolution (2018, methodology): “In partially observable environments, we can implement M through a recurrent neural network (RNN) [49, 74, 75, 78] to allow for better predictions based on memories of previous observation sequences.”
Revisiting the Arcade Learning Environment: Evaluation Protocols and Open Problems for General Agents (2017, background): “Additional components of the algorithm include clipping the rewards (as described above) and the use of experience replay (Lin, 1993) to decorrelate observations.”
Massively Parallel Methods for Deep Reinforcement Learning (2015, neutral): “First, DQN uses experience replay (Lin, 1993).”
Dueling Network Architectures for Deep Reinforcement Learning (2015, background): “Another key ingredient behind the success of DQN is experience replay (Lin, 1993; Mnih et al., 2015).”
A Deep Reinforcement Learning Chatbot (2017, methodology): “We use Q-learning with experience replay to learn the policy, since it is simple and has been shown to be effective with policies parametrized by neural networks (Mnih et al. 2013, Lin 1993).”
Deep learning in neural networks: An overview (2014, methodology): “More recently, a rather deep GPU-CNN was used in a traditional RL framework to play several Atari 2600 computer games directly from 84 × 84 pixel 60 Hz video input (Mnih et al., 2013), using experience replay (Lin, 1993), extending previous work on Neural Fitted Q-Learning (NFQ) (Riedmiller, 2005).”
Designing Neural Network Architectures using Reinforcement Learning (2016, methodology): “The Q -value update is applied to the transitions in temporally reversed order, which has been shown to speed up Q -values convergence (Lin, 1993).”
RUDDER: Return Decomposition for Delayed Rewards (2018, background): “If RUDDER’s safe exploration discovers an episode with unseen delayed rewards, it is secured in a lessons replay buffer [20].”
Graying the black box: Understanding DQNs (2016, methodology): “The Deep Q-Network (DQN) algorithm (Mnih et al., 2015) increased training stability by introducing the target network and by using Experi- * These authors have contributed equally ence Replay (ER) (Lin, 1993).”
Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models (2018, methodology): “Parametric function approximators have also been used extensively in MBRL [21], [22], [23], [24], but”
Deep Generative Dual Memory Network for Continual Learning (2017, methodology): “t all the tasks jointly. 3 DEEP GENERATIVE DUAL MEMORY NETWORK The idea of replaying experience to a neural network dates back to the 1990s and has been used for various reinforcement learning tasks (Lin, 1993; Mnih et al., 2015). A study by O’Neill et al. (2010) suggests that experience replays occ”
Born to Learn: the Inspiration, Progress, and Future of Evolved Plastic Artificial Neural Networks (2017, background): “orcement Learning (RL) (Sutton and Barto, 1998). RL in neural networks (Pennartz, 1997) makes use of robotic or agent-environment problems with a reward signal that affects plasticity in the network (Lin, 1993; Niv et al., 2002; Soltoggio and Stanley, 2012; Auerbach et al., 2014; Pugh et al., 2014) ”
A Deep Hierarchical Approach to Lifelong Learning in Minecraft (2016, background): “(2) Skill - Experience Replay: We extend the regular ER (Lin, 1993) to incorporate skills and term this the Skill Experience Replay (S-ER).”
The Challenges of Continuous Self-Supervised Learning (2022, positive): “This idea is inspired by experience replay [39] commonly used in reinforcement learning [3, 49, 72] and supervised continual learning [31, 68].”