Adam H. Marblestone
Research Subjects
Adam H. Marblestone has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Edgar A. Duenez-Guzman
Google DeepMind
1 shared publication - Edward Hughes
Google DeepMind, London, UK
1 shared publication - Iain Dunning
London DeepMind London DeepMind London DeepMind London London DeepMind London DeepMind London DeepMind London DeepMind London UK
1 shared publication - Joel Z. Leibo
King's College London
1 shared publication - Julien Perolat
Princeton University ORFE
1 shared publication - Peter Sunehag
Google DeepMind London UK
1 shared publication
Latest Publications
- Open Access
Malthusian Reinforcement Learning
Here we explore a new algorithmic framework for multi-agent reinforcement learning, called Malthusian reinforcement learning, which extends self-play to include fitness-linked population size dynamics that drive ongoing innovation. In Malthusian RL, increases in a subpopulation's average return drive subsequent increases in its size, just as Thomas Malthus argued in 1798 was the relationship between preindustrial income levels and population growth. Malthusian reinforcement learning harnesses the competitive pressures arising from growing and shrinking population size to drive agents to explore regions of state and policy spaces that they could not otherwise reach. Furthermore, in environments where there are potential gains from specialization and division of labor, we show that Malthusian reinforcement learning is better positioned to take advantage of such synergies than algorithms based on self-play.
Open AccessMalthusian Reinforcement Learning
Here we explore a new algorithmic framework for multi-agent reinforcement learning, called Malthusian reinforcement learning, which extends self-play to include fitness-linked population size dynamics that drive ongoing innovation. In Malthusian RL, increases in a subpopulation's average return drive subsequent increases in its size, just as Thomas Malthus argued in 1798 was the relationship between preindustrial income levels and population growth. Malthusian reinforcement learning harnesses the competitive pressures arising from growing and shrinking population size to drive agents to explore regions of state and policy spaces that they could not otherwise reach. Furthermore, in environments where there are potential gains from specialization and division of labor, we show that Malthusian reinforcement learning is better positioned to take advantage of such synergies than algorithms based on self-play.