Refine
Keywords
- credit assignment (1)
- learning (1)
- modules (1)
- reinforcement (1)
- reward (1)
Institute
-
Credit assignment in multiple goal embodied visuomotor behavior
(2010)
- The intrinsic complexity of the brain can lead one to set aside issues related to its relationships with the body, but the field of embodied cognition emphasizes that understanding brain function at the system level requires one to address the role of the brain-body interface. It has only recently been appreciated that this interface performs huge amounts of computation that does not have to be repeated by the brain, and thus affords the brain great simplifications in its representations. In effect the brain’s abstract states can refer to coded representations of the world created by the body. But even if the brain can communicate with the world through abstractions, the severe speed limitations in its neural circuitry mean that vast amounts of indexing must be performed during development so that appropriate behavioral responses can be rapidly accessed. One way this could happen would be if the brain used a decomposition whereby behavioral primitives could be quickly accessed and combined. This realization motivates our study of independent sensorimotor task solvers, which we call modules, in directing behavior. The issue we focus on herein is how an embodied agent can learn to calibrate such individual visuomotor modules while pursuing multiple goals. The biologically plausible standard for module programming is that of reinforcement given during exploration of the environment. However this formulation contains a substantial issue when sensorimotor modules are used in combination: The credit for their overall performance must be divided amongst them. We show that this problem can be solved and that diverse task combinations are beneficial in learning and not a complication, as usually assumed. Our simulations show that fast algorithms are available that allot credit correctly and are insensitive to measurement noise.
-
Infants in control: Rapid anticipation of action outcomes in a gaze-contingent paradigm
(2012)
- Infants' poor motor abilities limit their interaction with their environment and render studying infant cognition notoriously difficult. Exceptions are eye movements, which reach high accuracy early, but generally do not allow manipulation of the physical environment. In this study, real-time eye tracking is used to put 6- and 8-month-old infants in direct control of their visual surroundings to study the fundamental problem of discovery of agency, i.e. the ability to infer that certain sensory events are caused by one's own actions. We demonstrate that infants quickly learn to perform eye movements to trigger the appearance of new stimuli and that they anticipate the consequences of their actions in as few as 3 trials. Our findings show that infants can rapidly discover new ways of controlling their environment. We suggest that gaze-contingent paradigms offer effective new ways for studying many aspects of infant learning and cognition in an interactive fashion and provide new opportunities for behavioral training and treatment in infants.
-
Bayesian cue integration as a developmental outcome of reward mediated learning
(2011)
- Average human behavior in cue combination tasks is well predicted by Bayesian inference models. As this capability is acquired over developmental timescales, the question arises, how it is learned. Here we investigated whether reward dependent learning, that is well established at the computational, behavioral, and neuronal levels, could contribute to this development. It is shown that a model free reinforcement learning algorithm can indeed learn to do cue integration, i.e. weight uncertain cues according to their respective reliabilities and even do so if reliabilities are changing. We also consider the case of causal inference where multimodal signals can originate from one or multiple separate objects and should not always be integrated. In this case, the learner is shown to develop a behavior that is closest to Bayesian model averaging. We conclude that reward mediated learning could be a driving force for the development of cue integration and causal inference.
