Skull-Closed Autonomous Development

Wang, Yuekai; Wu, Xiaofeng; Weng, Juyang

doi:10.1007/978-3-642-24955-6_25

Cited by 3 publications

(2 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The reconfiguration of system parameters by a human expert should be avoided during an agent's run time. Agents that can autonomously adjust their parameters to learn new tasks are also characterized as skull-closed (Wang et al 2011). This means that there is no need to stop the agent upon a problem, to open its settings, i.e.…”

Section: Motivationmentioning

confidence: 99%

EO-MTRNN: evolutionary optimization of hyperparameters for a neuro-inspired computational model of spatiotemporal learning

Wieser

Cheng

2020

Biol Cybern

View full text Add to dashboard Cite

For spatiotemporal learning with neural networks, hyperparameters are often set manually by a human expert. This is especially the case with multiple timescale networks that require a careful setting of the values of timescales in order to learn spatiotemporal data. However, this implies a cumbersome trial-and-error process until suitable parameters are found and it reduces the long-term autonomy of artificial agents, such as robots that are controlled by multiple timescale networks. To solve the problem, we propose the evolutionary optimized multiple timescale recurrent neural network (EO-MTRNN) that is inspired by the neural plasticity of the human cortex. Our proposed network uses a method of evolutionary optimization to adjust its timescales and to rewire itself in terms of number of neurons and synapses. Moreover, it does not require additional neural networks for pre-and postprocessing input-output data. We validate our EO-MTRNN by applying it to a proposed benchmark training dataset with single and multiple sequence training cases, as well as by applying it to sensory-motor data from a robot. We compare different configuration modes of the network, and we compare the learning performance between a network configuration with manually set hyperparameters and a configuration with automatically estimated hyperparameters. The results show that automatically estimated hyperparameters yield approximately 43% better performance than manually estimated ones, without overfitting the given teaching data. We also validate the generalization ability by successfully learning data that were not included in the hyperparameter estimation process.

show abstract

Section: Motivationmentioning

confidence: 99%

EO-MTRNN: evolutionary optimization of hyperparameters for a neuro-inspired computational model of spatiotemporal learning

Wieser

Cheng

2020

Biol Cybern

View full text Add to dashboard Cite

show abstract

“…WWN-5 is capable of detecting and recognizing the objects with different scale in the complex environments [15]. WWN-6 improves the architecture and mechanisms of the network according to the concept "skull closed" [16]. In this paper, the skull closed network WWN-6 is taught to learn a new task, looking for the object in the unknown natural backgrounds and recognizing it like a child.…”

Section: Introductionmentioning

confidence: 99%

Skull-closed autonomous development: WWN-6 using natural video

Wang

Weng

2012

The 2012 International Joint Conference on Neural Networks (IJCNN)

Self Cite

View full text Add to dashboard Cite

Abstract-While a physical environment interacts with a human individual through the brain's sensors and effectors, internal representations inside the skull-closed brain autonomously emerge and adapt throughout the lifetime. By "skull-closed", we mean that the brain inside the skull is off limit to all teachers in the external physical environment, except the brain's sensory ends and motor ends. We present the Where-What Network 6 (WWN-6), which has realized our goal of fully autonomous development inside a closed network "skull". This means that the human programmer is not allowed to handcraft the internal representation for any fixed extra-body concepts. For example, the meanings of specific values of location or type concept are not known during the programming time. Such meanings emerge through associations imbedded in 'postnatal" experience. This capability is especially challenging when one considers the fact that most elements in the sensory ends are irrelevant to the signals at the effector ends (e.g., many background pixels). How does each vector value in the effector find its corresponding pattern in the correct patch of the sensory image? We outline this autonomous learning theory for the brain and present how the developmental program (DP) of WWN-6 enables the network to perform for attending and recognizing objects in complex backgrounds using natural video. The inputs to the agent (i.e., the network) are not artificially synthesized images as WWNs used before, but drawn from continuous video taken from natural settings where, in general, everything is moving.

show abstract

Skull-Closed Autonomous Development: Object-Wise Incremental Learning

Wang

Weng

2012

Advances in Neural Networks – ISNN 2012

View full text Add to dashboard Cite

Skull-Closed Autonomous Development

Cited by 3 publications

References 6 publications

EO-MTRNN: evolutionary optimization of hyperparameters for a neuro-inspired computational model of spatiotemporal learning

EO-MTRNN: evolutionary optimization of hyperparameters for a neuro-inspired computational model of spatiotemporal learning

Skull-closed autonomous development: WWN-6 using natural video

Skull-Closed Autonomous Development: Object-Wise Incremental Learning

Contact Info

Product

Resources

About