ioeu

What fields would you guide them to instead?

I may be one of the developers you speak of (with academic aspirations), presently considering my path forward.

I'm sceptical if going down the ML swamp is the best way forward.

In retrospect that seems much simpler than what I proposed in my edit, thanks.

> For each image, the algorithm produces a heat map showing which parts of the image are most memorable. By emphasizing different regions, they can potentially increase the image’s memorability.

Most memorable, according to human subjects subjective thought on the matter?

> The team then pitted its algorithm against human subjects by having the model predicting how memorable a group of people would find a new never-before-seen image. It performed 30 percent better than existing algorithms and was within a few percentage points of the average human performance.

Who's to say human subjects are any good at objectively judging how memorable a photo is? I feel like I'm missing something.

Edit: Riight, I guess it could be based on observing neural activity in human subjects while they look at photos. That makes a lot more sense.

> While deep-learning has propelled much progress in object recognition and scene understanding, predicting human memory has often been viewed as a higher-level cognitive process that computer scientists will never be able to tackle

Seriously?

Has it seriously often been viewed like that?

Your response is totally understandable given my (very clumsy) comment. Let me make myself more clear; I'm in the process of reading up on what material one should study to set oneself up to ultimately be able to consider AGI research as a career choice.

You say "mathematics, cognitive science, and computer science at increasing levels of sophistication.". That's probably true, I'd recon a fair bit of biology, chemistry and physics is most likely also needed.

My question, though, is what specific branches of said subjects?

I totally agree! If anybody knows of something similar for artificial general intelligence I'm all ears, currently in the process of reading up on it.

Did you look for any similar software before you started building? Did you find anything?

I'm inclined to agree, I think it should work for most basic types using controls for forward/rewind and optionally only looking at a subset of the data.

A potential problem with visualizing more complex structures may be that people's mental representations for them probably differ, making it really hard to offer a visualization that maps to each person's mental model of each specific (complex) data structure.

What do you mean by: "There are also controls you can't see that make hierarchically navigating from groups of data structures to single data structures to single elements, easy."?

I'm happy we came across then! I find this type of visualiation to be intriguing, and I may give it a stab in the future.

We should definitely keep in touch.

Thanks! I'll hopefully have a look, at some point.

I would've enjoyed reading more about your approach seeing as it seems to be very alike to what I wanted to try and solve. I wanted to, similar to what you describe, be able to visualize the program's data structures as I visualize them in my head avoiding writing one-off, text-based, sub-optimal printing code.

I'm still interested in the idea (i.e. visualizing a program by its state) although I fear it may not be practical for non-trivial programs with complex, large and rapidly changing data structures.