Limits...
Neural modularity helps organisms evolve to learn new skills without forgetting old skills.

Ellefsen KO, Mouret JB, Clune J - PLoS Comput. Biol. (2015)

Bottom Line: Modularity intuitively should reduce learning interference between tasks by separating functionality into physically distinct modules in which learning can be selectively turned on or off.To produce modularity, we evolve neural networks with a cost for neural connections.We show that this connection cost technique causes modularity, confirming a previous result, and that such sparsely connected, modular networks have higher overall performance because they learn new skills faster while retaining old skills more and because they have a separate reinforcement learning module.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer and Information Science, Norwegian University of Science and Technology, Trondheim, Norway.

ABSTRACT
A long-standing goal in artificial intelligence is creating agents that can learn a variety of different skills for different problems. In the artificial intelligence subfield of neural networks, a barrier to that goal is that when agents learn a new skill they typically do so by losing previously acquired skills, a problem called catastrophic forgetting. That occurs because, to learn the new task, neural learning algorithms change connections that encode previously acquired skills. How networks are organized critically affects their learning dynamics. In this paper, we test whether catastrophic forgetting can be reduced by evolving modular neural networks. Modularity intuitively should reduce learning interference between tasks by separating functionality into physically distinct modules in which learning can be selectively turned on or off. Modularity can further improve learning by having a reinforcement learning module separate from sensory processing modules, allowing learning to happen only in response to a positive or negative reward. In this paper, learning takes place via neuromodulation, which allows agents to selectively change the rate of learning for each neural connection based on environmental stimuli (e.g. to alter learning in specific locations based on the task at hand). To produce modularity, we evolve neural networks with a cost for neural connections. We show that this connection cost technique causes modularity, confirming a previous result, and that such sparsely connected, modular networks have higher overall performance because they learn new skills faster while retaining old skills more and because they have a separate reinforcement learning module. Our results suggest (1) that encouraging modularity in neural networks may help us overcome the long-standing barrier of networks that cannot learn new skills without forgetting old ones, and (2) that one benefit of the modularity ubiquitous in the brains of natural animals might be to alleviate the problem of catastrophic forgetting.

No MeSH data available.


Related in: MedlinePlus

Comparing the retention and forgetting of networks from the two treatments.P&CC networks, which are more modular, are better at retaining associations learned on a previous task (winter associations) while learning a new task (summer associations), better at learning new (summer) associations, and significantly better when measuring performance on both the associations for the original task (winter) and the new task (summer). Note that networks were evolved with five days per season, so the results during those first five days are the most informative regarding the evolutionary mitigation of catastrophic forgetting: we show additional days to reveal longer-term consequences of the evolved architectures. Solid lines show median performance and shaded areas indicate 95% bootstrapped confidence intervals of the median. The retention scores (left panel) are normalized relative to the original performance before training on the new task (an unnormalized version is provided as Supp. S6 Fig). During all performance measurements, learning was disabled to prevent such measurements from changing an individual’s known associations (Methods).
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4383335&req=5

pcbi.1004128.g008: Comparing the retention and forgetting of networks from the two treatments.P&CC networks, which are more modular, are better at retaining associations learned on a previous task (winter associations) while learning a new task (summer associations), better at learning new (summer) associations, and significantly better when measuring performance on both the associations for the original task (winter) and the new task (summer). Note that networks were evolved with five days per season, so the results during those first five days are the most informative regarding the evolutionary mitigation of catastrophic forgetting: we show additional days to reveal longer-term consequences of the evolved architectures. Solid lines show median performance and shaded areas indicate 95% bootstrapped confidence intervals of the median. The retention scores (left panel) are normalized relative to the original performance before training on the new task (an unnormalized version is provided as Supp. S6 Fig). During all performance measurements, learning was disabled to prevent such measurements from changing an individual’s known associations (Methods).

Mentions: P&CC networks retain higher performance on the original task when learning a new task (Fig. 8, left). They also learn the new task better (Fig. 8, center). The combined effect significantly improves performance (Fig. 8, right), meaning P&CC networks are significantly better at learning associations in a new season while retaining associations from a previous one.


Neural modularity helps organisms evolve to learn new skills without forgetting old skills.

Ellefsen KO, Mouret JB, Clune J - PLoS Comput. Biol. (2015)

Comparing the retention and forgetting of networks from the two treatments.P&CC networks, which are more modular, are better at retaining associations learned on a previous task (winter associations) while learning a new task (summer associations), better at learning new (summer) associations, and significantly better when measuring performance on both the associations for the original task (winter) and the new task (summer). Note that networks were evolved with five days per season, so the results during those first five days are the most informative regarding the evolutionary mitigation of catastrophic forgetting: we show additional days to reveal longer-term consequences of the evolved architectures. Solid lines show median performance and shaded areas indicate 95% bootstrapped confidence intervals of the median. The retention scores (left panel) are normalized relative to the original performance before training on the new task (an unnormalized version is provided as Supp. S6 Fig). During all performance measurements, learning was disabled to prevent such measurements from changing an individual’s known associations (Methods).
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4383335&req=5

pcbi.1004128.g008: Comparing the retention and forgetting of networks from the two treatments.P&CC networks, which are more modular, are better at retaining associations learned on a previous task (winter associations) while learning a new task (summer associations), better at learning new (summer) associations, and significantly better when measuring performance on both the associations for the original task (winter) and the new task (summer). Note that networks were evolved with five days per season, so the results during those first five days are the most informative regarding the evolutionary mitigation of catastrophic forgetting: we show additional days to reveal longer-term consequences of the evolved architectures. Solid lines show median performance and shaded areas indicate 95% bootstrapped confidence intervals of the median. The retention scores (left panel) are normalized relative to the original performance before training on the new task (an unnormalized version is provided as Supp. S6 Fig). During all performance measurements, learning was disabled to prevent such measurements from changing an individual’s known associations (Methods).
Mentions: P&CC networks retain higher performance on the original task when learning a new task (Fig. 8, left). They also learn the new task better (Fig. 8, center). The combined effect significantly improves performance (Fig. 8, right), meaning P&CC networks are significantly better at learning associations in a new season while retaining associations from a previous one.

Bottom Line: Modularity intuitively should reduce learning interference between tasks by separating functionality into physically distinct modules in which learning can be selectively turned on or off.To produce modularity, we evolve neural networks with a cost for neural connections.We show that this connection cost technique causes modularity, confirming a previous result, and that such sparsely connected, modular networks have higher overall performance because they learn new skills faster while retaining old skills more and because they have a separate reinforcement learning module.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer and Information Science, Norwegian University of Science and Technology, Trondheim, Norway.

ABSTRACT
A long-standing goal in artificial intelligence is creating agents that can learn a variety of different skills for different problems. In the artificial intelligence subfield of neural networks, a barrier to that goal is that when agents learn a new skill they typically do so by losing previously acquired skills, a problem called catastrophic forgetting. That occurs because, to learn the new task, neural learning algorithms change connections that encode previously acquired skills. How networks are organized critically affects their learning dynamics. In this paper, we test whether catastrophic forgetting can be reduced by evolving modular neural networks. Modularity intuitively should reduce learning interference between tasks by separating functionality into physically distinct modules in which learning can be selectively turned on or off. Modularity can further improve learning by having a reinforcement learning module separate from sensory processing modules, allowing learning to happen only in response to a positive or negative reward. In this paper, learning takes place via neuromodulation, which allows agents to selectively change the rate of learning for each neural connection based on environmental stimuli (e.g. to alter learning in specific locations based on the task at hand). To produce modularity, we evolve neural networks with a cost for neural connections. We show that this connection cost technique causes modularity, confirming a previous result, and that such sparsely connected, modular networks have higher overall performance because they learn new skills faster while retaining old skills more and because they have a separate reinforcement learning module. Our results suggest (1) that encouraging modularity in neural networks may help us overcome the long-standing barrier of networks that cannot learn new skills without forgetting old ones, and (2) that one benefit of the modularity ubiquitous in the brains of natural animals might be to alleviate the problem of catastrophic forgetting.

No MeSH data available.


Related in: MedlinePlus