Submission Category : Applications , Preference : ORAL Reinforcement Learning for Spoken Dialogue Systems

By Satinder Singh, Michael Kearns, Diane J. Litman, Marilyn A. Walker. 2007

Subfields: Natural Language Processing, Reinforcement Learning

How researchers see this paper

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

In Natural Language Processing

In Reinforcement Learning

Citations in the corpus

Open-Domain Conversational Agents: Current Progress, Open Problems, and Future Directions (2020, background): “Classical goaloriented dialogue literature typically uses structured knowledge, slot filling or labeling, and studies reinforcement learning extensively (Singh et al., 2000).”
Safe Policy Improvement with Soft Baseline Bootstrapping (2019, background): “…that naturally fit a setting where data collection and optimization are decoupled (contrary to online learning which integrates the two): e.g. dialogue systems (Singh et al., 1999), technical process control (Ernst et al., 2005; Riedmiller, 2005), medical applications (Guez et al., 2008).”
Safe Policy Improvement with Baseline Bootstrapping (2017, background): “dialogue systems (Singh et al., 1999), technical process control (Ernst et al.”
GuessWhat?! Visual Object Discovery through Multi-modal Dialogue (2016, background): “A promising alternative is goal-directed dialogue tasks [29, 38, 42, 41] where agents converse to pursue a goal rather than casually chit-chat.”
Deep Reinforcement Learning with a Natural Language Action Space (2015, neutral): “uage processing, reinforcement learning has been applied to a dialogue management system that converses with a human user by taking actions that generate natural language (Schefﬂer & Young, 2002; Singh et al., 1999). There has also been interest in extracting textual knowledge to improve game co”
Hybrid Reinforcement/Supervised Learning of Dialogue Policies from Fixed Data Sets (2008, background): “Task of each system utterance (Walker and Passonneau 2001; Walker, Passonneau, and Boland 2001). In addition, the results of user questionnaires are available, but only for the 2001 corpus. Table 1 shows some statistics for the two collections. In the 2000 collection each turn contains only one utte”
BBQ-Networks: Efficient Exploration in Deep Reinforcement Learning for Task-Oriented Dialogue Systems (2016, background): “…work by perturbing the reward function to add a bonus for uncovering surprising transitions, i.e., state transitions given low probability by a dynamics model, or previously rarely seen states (Stadie, Levine, and Abbeel 2015; Houthooft et al. 2016a; Houthooft et al. 2016b; Bellemare et al. 2016).”
End-to-end optimization of goal-driven and visually grounded dialogue systems (2017, methodology): “As a consequence, reinforcement learning has been applied to dialogue systems since the late 90s [Levin et al., 1997; Singh et al., 1999] and dialogue optimization has been generally more studied than dialogue generation.”
Situated Dialogue Learning through Procedural Environment Generation (2021, methodology): “Goal oriented Dialogue Sub-tasks within the overall task of goal oriented dialogue, such as dialogue state management (Singh et al., 2000; Pietquin et al., 2011; Fatemi et al., 2016) and response generation (Li et al., 2016) have used RL to boost performance.”
Score-based Inverse Reinforcement Learning (2016, background): “INTRODUCTION Reinforcement Learning (RL) has seen a growing interest this last decade both from a theoretical and a practical point of view, especially in the domain of Human-Machine Interaction (HMI) where it is now part of the state of the art [20, 13, 24].”
Sample-efficient batch reinforcement learning for dialogue management optimization (2011, methodology): “…there has been an important focus on machine learning methods for automatic learning of dialogue management strategies since the late 90s [Levin and Pieraccini 1998; Singh et al. 1999; Levin et al. 2000; Schef.er and Young 2001; Pietquin 2004; Williams and Young 2007; Lemon and Pietquin 2007].”
Sequence-to-Sequence Asr Optimization Via Reinforcement Learning (2017, background): “Many reinforcement learning applications exist, including building an agent that can learn how to play a game without any explicit knowledge [22, 23], control tasks in robotics [24], and dialogue system agents [25, 26].”
POMDP-Based Statistical Spoken Dialog Systems: A Review (2013, methodology): “The weights can be inferred from the annotated corpus using regression, and this regression can then be used as the reward function in reinforcement learning [129], [130].”
Calibrated Model-Based Deep Reinforcement Learning (2019, background): “, 2018), dialogue systems (Singh et al., 2000), education (Rollinson & Brunskill, 2015), scientific discovery (McIntire et al.”