Tag Archives: Prediction By Simulation

Doing a more intelligent exploration in RL based on measuring uncertainty through prediction

Xiaoshu Zhou, Fei Zhu, Peiyao Zhao, Within the scope of prediction: Shaping intrinsic rewards via evaluating uncertainty, Expert Systems with Applications, Volume 206, 2022 DOI: 10.1016/j.eswa.2022.117775.

The agent of reinforcement learning based approaches needs to explore to learn more about the environment to seek optimal policy. However, simply increasing the frequency of stochastic exploration sometimes fails to work or even causes the agent to fall into traps. To solve the problem, it is essential to improve the quality of exploration. An approach, referred to as the scope of prediction based on uncertainty exploration (SPE), is proposed, taking advantage of the uncertainty mechanism and considering the stochasticity of prospecting. As by uncertainty mechanism, the unexpected states make more curiosity, the model derives higher uncertainty by projecting future scenarios to compare with the actual future to explore the world. The SPE method utilizes a prediction network to predict subsequent observations and calculates the mean squared difference value of the real observations and the following observations to measure uncertainty, encouraging the agent to explore unknown regions more effectively. Moreover, to reduce the noise interference caused by uncertainty, a reward-penalty model is developed to discriminate the noise by current observations and action prediction for future rewards to improve the interference ability against noise so that the agent can escape from the noisy region. Experiment results showed that deep reinforcement learning approaches equipped with SPE demonstrated significant improvements in simulated environments.

On the role and limitations of motor internal simulation as a way of predicting the effects of a future action in the brain

Myrthel Dogge, Ruud Custers, Henk Aarts, Moving Forward: On the Limits of Motor-Based Forward Models. Trends in Cognitive Sciences, Volume 23, Issue 9, 2019, Pages 743-753, DOI: 10.1016/j.tics.2019.06.008.

The human ability to anticipate the consequences that result from action is an essential building block for cognitive, emotional, and social functioning. A dominant view is that this faculty is based on motor predictions, in which a forward model uses a copy of the motor command to predict imminent sensory action-consequences. Although this account was originally conceived to explain the processing of action-outcomes that are tightly coupled to bodily movements, it has been increasingly extrapolated to effects beyond the body. Here, we critically evaluate this generalization and argue that, although there is ample evidence for the role of predictions in the processing of environment-related action-outcomes, there is hitherto little reason to assume that these predictions result from motor-based forward models.