Olivier Georgeon's research blog—also known as the story of little Ernest, the developmental agent.

Keywords: situated cognition, constructivist learning, intrinsic motivation, bottom-up self-programming, individuation, theory of enaction, developmental learning, artificial sense-making, biologically inspired cognitive architectures, agnostic agents (without ontological assumptions about the environment).

Friday, October 16, 2009

Iconic visual perception

Ernest 7.4 has a basic iconic visual system implemented. He can now perceive his surrounding environment under the form of an icon made of three parts: [left, forward, right]. These parts correspond to the three squares around Ernest. This icon is displayed in the upper-right corner of the video.

This environment offers 5 different possible icons. Ernest must learn them and simultaneously learn what behavior is the most satisfying in the context of each of them.

This video shows that this learning is pretty fast. With this iconic mechanism, Ernest succeeds much better in this environment than with the sequential learning mechanism only. Indeed, this task is more tailored to visual learning than to sequential learning, because all the information needed to make a choice is easily accessible in the visual field.

In this experiment, the "visual" perception is not implemented as sensory-motor schemas. The visual perception rather follows a second parallel process of learning and recollection. New learned icons are stored in Ernest's iconic memory; and new or recollected icons are activated in Ernest's short-term memory. As an element of Ernest's short-term memory, the currently-percieved icon is part of the context of the new learned schemas.

Ernest 7.4 has three primitive schemas with the following settings:
- [move forward, succeed, 10] Ernest enjoys moving forward.
- [move forward, fail, -10] Ernest dislikes bumping walls.
- [turn left or right, succeed, 0] Ernest is indifferent of turning toward an empty square.
- [turn left or right, fail, -5] Ernest dislikes turning toward a wall.


Dran said...

Is there any satisfaction attached to visual stimulus?

Olivier said...

Hello Dran,
No there is no satisfaction directly attached to iconic perception.
The satisfaction is only attached to the primitive acts as explained in the post.
Iconic perception is however attached to the schemas' context and that's how iconic perception plays a role in the choice of behavior.
I know the implementation is not detailed enough but it is only exploratory so far, and it will certainly evolve in future developments.

Dran said...

OK. I was just thinking about how there was satisfaction connected with touching. Touching is clearly more of an action than seeing though.

It seems like giving the agent iconic perception makes its life much easier, which actually makes the whole scenario less interesting. There is no delay between action and resulting satisfaction, so that the model no longer demonstrates the agent's abilities as well as with touch-based perception.

I'm Daniel Demski from the Soar workshop, by the way.

Olivier said...

Hello Daniel,
I am glad to see you again here. Yes you are perfectly right, the scenario is much less interesting now that the agent has these new perceptual skills. The purpose here is only to prepare the agent for future more complex experiments.