TAILIEUCHUNG - Báo cáo khoa học: "Reinforcement Learning for Mapping Instructions to Actions"

In this paper, we present a reinforcement learning approach for mapping natural language instructions to sequences of executable actions. We assume access to a reward function that defines the quality of the executed actions. During training, the learner repeatedly constructs action sequences for a set of documents, executes those actions, and observes the resulting reward. We use a policy gradient algorithm to estimate the parameters of a log-linear model for action selection. We apply our method to interpret instructions in two domains — Windows troubleshooting guides and game tutorials. . | Reinforcement Learning for Mapping Instructions to Actions . Branavan Harr Chen Luke S. Zettlemoyer Regina Barzilay Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology branavan harr Isz regina @ Abstract In this paper we present a reinforcement learning approach for mapping natural language instructions to sequences of executable actions. We assume access to a reward function that defines the quality of the executed actions. During training the learner repeatedly constructs action sequences for a set of documents executes those actions and observes the resulting reward. We use a policy gradient algorithm to estimate the parameters of a log-linear model for action selection. We apply our method to interpret instructions in two domains Windows troubleshooting guides and game tutorials. Our results demonstrate that this method can rival supervised learning techniques while requiring few or no annotated training 1 Introduction The problem of interpreting instructions written in natural language has been widely studied since the early days of artificial intelligence Winograd 1972 Di Eugenio 1992 . Mapping instructions to a sequence of executable actions would enable the automation of tasks that currently require human participation. Examples include configuring software based on how-to guides and operating simulators using instruction manuals. In this paper we present a reinforcement learning framework for inducing mappings from text to actions without the need for annotated training examples. For concreteness consider instructions from a Windows troubleshooting guide on deleting temporary folders shown in Figure 1. We aim to map 1Code data and annotations used in this work are available at http rbg code rl o Click start point to search and then click for files or folders. o In the search results dialog box on the tools menu click folder options. o In the folder options dialog .

• Báo cáo khoa học
• Reinforcement Learning
• Mapping Instructions to Actions
• S
• Branavan
• báo cáo khoa học
• báo cáo ngôn ngữ
• xử lý ngôn ngữ tự nhiên
• Machine learning
• Lecture Machine learning
• Machine learning algorithms
• Data analysis methods
• Delayed rewards
• Information Sciences
• Towards safe reinforcement learning
• Industrial grid warehousing
• Markov decision processes
• Safe reinforcement learning
• Journal of Automation and Control Engineering
• Efficient transfer learning
• Reinforcement learning using sparse coding
• Normal Q learning
• Means of sparse coding
• Proposed method significantly
• Forecasting Systems
• System Identification
• Controller Design
• Grammatical Inference
• Q Learning
• Multi agent system
• Distributed system
• Markov decision process
• Factored Markov decision process
• Electronic Engineering
• Balancing control
• Hierarchical reinforcement learning
• Feature value vector
• Hierarchical Reinforcement Learning and Hidden Markov Models
• Task Oriented Natura
• l Language Generation
• ngôn ngữ tự nhiên
• From Structured Prediction
• Inverse Reinforcement Learning
• Hal Daum
• An ISU Dialogue System Exhibiting Reinforcement Learning
• Dialogue Policies
• Generic Slot filling
• TALK In car System
• scientific reports
• model language
• process natural language
• A Comparative Study
• Reinforcement Learning Techniques
• Dialogue Management
• Using Reinforcement Learning
• Better Model
• Dialogue State
• Decreasing induction motor loss using reinforcement learning
• Decreasing induction motor loss
• The motor works
• The standard driver of motor
• International Journal of Computer Networks and Communications Security
• Improve the routing algorithm
• Wireless sensor networks
• Reinforcement learning strategy
• The semi processed data
• Renewable energy
• Three phase inverter
• Volt age control
• Reinforcement learning algorithm
• Vehicular Mobile Networks
• Energy optimization
• Modern life
• Internet of Things
• Recommendation system
• Personalized web search
• Cooperative learning
• Unsupervised learning
• Theoretical neuroscience
• Annotated bibliography
• Network models
• plasticity and learning
• Classical conditioning
• Representational learning
• Journal of Computer Science and Cybernetics
• Efficient navigation framework
• Autonomous mobile robots
• The mobile robots
• báo cáo hóa học
• công trình nghiên cứu về hóa học
• tài liệu về hóa học
• cách trình bày báo cáo
• Energy Management
• Buildings Using MATLAB
• Control Applications
• Synchronous Generator
• Advanced Control
• Strategies Simulation
• Humanoid Robotic
• Virtual Reality Simulation
• Humanoid Biped
• Connectives Acquisition
• Interactive Robots
• Human Arm