I speculate that there exists something like ”recognition improvement”. Similar to noise reduction but with a wider usage.
Bad handwriting. Like mine for instance. Is something I don’t want. Sometimes if I don’t publish within a certain time period I forget a little what the idea was that I had written down.
Then I wonder. Is there a relation between complexity and recognition ability? Like bad handwriting becomes hard to read after some time.
Is there then a purpose to an aging memory?
Could you simulate an aging memory and see how well it can be used to recognize the object. The idea is to make changes to the object to withstand an aging memory.
Like the traveling salesman problem. You don’t want the thought to travel a long distance between memories.
This gives me the idea that a aging memory is feature of machine learning.
To make use of the resources we have.