Relational Instance Based Regression for Relational Reinforcement Learning

By Kurt Driessens, J. Ramon. 2003

85 citations recorded by the source corpus.

How researchers see this paper

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

Citations in the corpus

Fitted Q-Learning for Relational Domains (2020, methodology): “This work was later extended by others (Driessens and Ramon 2003; Driessens and Dˇzeroski 2004; Driessens, Ramon, and Blockeel 2001; G¨artner, Driessens, and Ramon 2003) w.r.t building incremental RRTs, providing guidance to the learner and regression algorithms using kernels and gaus-sian…”
Relational Deep Reinforcement Learning (2018, background): “The core idea behind RRL is to combine reinforcement learning with relational learning or Inductive Logic Programming [16] by representing states, actions and policies using a first order (or relational) language [8, 9, 17, 18].”
Planning in Discrete and Continuous Markov Decision Processes by Probabilistic Programming (2015, unknown): “[3,5,26]) for model-free reinforcement learning.”
Neural Logic Reinforcement Learning (2019, background): “There are variants of this work (Driessens & Ramon, 2003; Driessens & Dˇzeroski, 2004) that extend the work, however, all these algorithms employ non-differential operations which makes it hard to apply new breakthroughs happened in DRL community.”
Relational Reinforcement Learning An Overview (2004, methodology): “Three regression algorithms have been developed for use in this RRL system: the TG algorithm, which i n crementally builds rst order regression trees, an instance based algorithm called RIB and a kernel based algorithm KBR that uses Gaussian processes as the regression technique.”
Integrating Guidance into Relational Reinforcement Learning (2004, methodology): “For the instance-based RRL-RIB we used the following distance definition for two (state, action) pairs in the blocks world: 1.”
Solving Relational and First-Order Logical Markov Decision Processes: A Survey (2012, methodology): “adaptive tree Q-learning BW, Tsume-Go RIB (Driessens and Ramon, 2003) IB Q-learning BW TRENDI (Driessens and Džeroski, 2005) tree + IB Q-learning BW KBR (Gärtner et al, 2003) kernels Q-learning BW MARLIE (Croonenborghs et al, 2007b) prob.”
Safe Exploration of State and Action Spaces in Reinforcement Learning (2012, background): “Forbes and Andres (2002) suggest the removal of cases that contribute least to the overall approximation, while Driessens and Ramon (2003) pursue a more error-oriented view and propose the deletion of cases that contribute most to the prediction error of other examples.”
Non-parametric policy gradients: a unified treatment of propositional and relational domains (2008, methodology): “…developed for use in this RL system that employ relational regression trees (Driessens et al., 2001), relational instance based regression (Driessens & Ramon, 2003), graph kernels and Gaussian processes (G¨artner et al., 2003) and relational model-trees (Driessens & D.zeroski, 2005).”
Learning to Manipulate Articulated Objects in Unstructured Environments Using a Grounded Relational Representation (2008, methodology): “In our case, this representation is instancebased [6].”
Graph kernels and Gaussian processes for relational reinforcement learning (2006, methodology): “Other instance selection strategies that might be useful are suggested in [7] and have there successfully been applied in instance based RRL.”
Relational reinforcement learning (2001, unknown): “The instance based RIB algorithm [4] uses the recent development of relational distances.”