Hierarchical Policy Gradient Algorithms

By Mohammad Ghavamzadeh, Sridhar Mahadevan. 2003

65 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

Deep learning in neural networks: An overview (2014, methodology): “An important class of DS methods for NNs are Policy Gradient methods (Williams, 1986, 1988, 1992a; Baxter and Bartlett, 1999; Sutton et al., 1999a; Aberdeen, 2003; Ghavamzadeh and Mahadevan, 2003; Kohl and Stone, 2004; Wierstra et al., 2007, 2008; Rückstieß et al., 2008; Peters and Schaal, 2008b,a; ”
Zero-Shot Task Generalization with Multi-Task Deep Reinforcement Learning (2017, background): “Typically these have the form of a meta controller and a set of lower-level controllers for subtasks (Sutton et al., 1999; Dietterich, 2000; Parr and Russell, 1997; Ghavamzadeh and Mahadevan, 2003; Konidaris et al., 2012; Konidaris and Barto, 2007).”
On Learning to Think: Algorithmic Information Theory for Novel Combinations of Reinforcement Learning Controllers and Recurrent Neural World Models (2015, methodology): “1, except that C is now an FNN or RNN trained by evolutionary algorithms [200, 255, 105, 56, 68] applied to NNs [165, 321, 180, 259, 72, 90, 89, 110, 94], or by policy gradient methods [314, 315, 316, 274, 18, 1, 63, 128, 313, 210, 192, 191, 256, 85, 312, 190, 82, 93][245, Sec.”
One Big Net For Everything (2018, methodology): “…from this prior knowledge, learning a new task through additional weight changes more quickly than learning the task from scratch, e.g., [17, 100, 13], ideally through optimal algorithmic transfer learning , like in the at least asymptotically Optimal Ordered Problem Solver [69], where new…”
Hierarchical multi-agent reinforcement learning (2001, methodology): “In this direction, we presented a family of HRL algorithms suitable for problems with continuous state and/or action spaces, using a mixture of policy gradient-based RL and value function-based RL methods [11].”
Hierarchical Approaches (2012, background): “Ghavamzadeh and Mahadevan (2003) use a hierarchical policy gradient approach to tackle reinforcement learning problems with continuous state and action spaces.”
Learning concurrent motor skills in versatile solution spaces (2012, background): “Using hierarchical policy representations, also called hierarchical RL, is a promising idea and has already been shown to accelerate learning in many situations [15], [16], [14].”
MT-Opt: Continuous Multi-Task Robotic Reinforcement Learning at Scale (2021, background): “…and composite skills has been addressed through hierarchical reinforcement learning with options [75, 7, 14], combining multiple sub-tasks [6, 16, 49, 26, 69, 89], reusing samples between tasks [39], relabeling experience in hindsight [3], introducing demonstrations [58, 81, 29, 41, 70, 66, 67].”