What other software language does the similar to Wolfram language to synthesizing with ChatGPT?

Awesome class and now understand the ChatGPT under the hood. It is really amazing that Steven Wolfram already knows the future and started using neural networks early. How did Steven get into using neural networks? Any good course that you recommend to learn about neural network?

Hi Ana! I could see that being a possibility. I think the most likely thing is that traffic cameras could recognize how many cars there are along each direction and what sorts of speeds they're seeing, then change the lights dependent on those inputs. There are traditional ways to do this measurement, but with neural nets you can just plug in any camera and a cheap circuit board and very quickly get reasonable estimates for these parameters (as well as, say, bikes and pedestrians). You could then use this data to determine when it's time to change the lights, in something like the following fashion:

• Camera uses neural nets to detect the number of cars, people, and cyclists, as well as their speeds
• This data is fed into a neural net which has been trained on data that looks like "if you have this many cars and people and cyclists and these are their speeds, switch the light"
• This decision is fed into some kind of traditional failsafe, as you couldn't guarantee with a neural net alone that you wouldn't wind up having an intersection have green lights in two perpendicular directions, which would be very bad!

So, you're right that you would need some kind of non-neural network "final check" for a case like this, and also right that this isn't something you'd need (or want) to use the newer generative-style AI for.

I think it could, but generally the problem of memorization is not too serious a one in modern neural net development, is my understanding. The ideal would be that the net builds a good model of what traffic flow looks like, and as long as there's a good amount of data it should be able to account for the fact that traffic has periods of extreme irregularity (e.g. crashes or construction). If the designers realize that it's doing less modeling and more memorizing, they can change the training to account for that (and typically, it's pretty easy to realize during the training process whether a net is "overfitting" or memorizing).

Of course! My background is in physics, so while this stuff is maybe a little closer to home, I was really impressed when I learned a bit about how it works and how "simple" everything is on the inside, mechanically speaking.

With models like these, the training process is something that costs O(billions) of dollars, so whenever you tell the model something in your prompt, it does not learn it or store it—it just takes it in as a token and passes it through the layers, and based on that alone it's able to provide a relevant answer. So the answer to your question is: only the current session that your student has open will be aware of the new info, and that's because a "session" stores all input and output and uses the whole dang thing as a big input every time.

That said—I can't guarantee that the things you type into a ChatGPT session are not being stored and used for further training. I'd be kind of surprised if they weren't, but that's just a technological judgment and not a moral or probabilistic one.

Thank you very much! I'm glad I was able to properly convey the ideas outlined in Stephen's book.

Hey Arben, I had an open session of 14.0 running on my Mac and I double clicked the Machine Learning notebook after downloading it; my Mathematica session went off to LaLa land. Wouldn't even obey the interrupt Cmd + Option+ [.] or CTRL C. Had to force quit from MacOS. I started a new session and opened the notebook from within Mathematica and it opened as I expected. But I have not "enabled dynamics"... It should have opened the same by double clicking on the file, I think, yes?

AI has so many ways to describe it: AI, Machine Learning, Deep Learning, and Generative AI. Which term is the best way to describe Wolfram's approach? Why is Wolfram's purpose to get into AI world?

They're all a little bit different:

• "Machine learning" is generally used to describe the process of training neural nets
• "Deep learning" is a subset of machine learning referring to the training of "deep" neural networks, i.e. those with many layers
• "AI" is super general and encompasses the above and more—it's so general that it can lead to miscommunications and mystification, though
• "Generative AI" refers to "AI" (usually based on neural nets which have been trained via machine learning) which can "generate" novel text, images, audio, and so on

As for what we're doing at Wolfram, we've supported neural nets and machine learning for a long time now, and our fully symbolic approach is really well-suited to the construction and understanding of neural nets. Our new LLM (one example of generative AI) functionality takes advantage of the benefits of this new technology to do things like create helpful personas, write code, check or rephrase writing, and so on inside notebooks. Alan will address this in more detail on Friday!

Hi everybody—I've just uploaded to the materials folder a slightly updated version of today's notebook with a few fixes we encountered during today's session.

Hey Carl! Thanks for the kind words for our materials :). While I'm not a machine learning pro, I think that your example use case is quite doable (and depending on the image resolution, you may need fewer images than you imagine—high resolution images by definition have a lot of information stored in them, and IME even training on the order of tens of examples can be sufficient to do binary categorization with high accuracy in those cases).

As for perspective transformations, that could be a little trickier. I'm not sure what the extent of the different angles and viewpoints might be, but it could be that your training data could include an image and various transformations of it created with something like ImagePerspectiveTransformation or similar?

For spotting differences, theoretically you could train a neural network to do that by providing training data consisting of two input images with the output being a selected region where differences occur, but I'm not sure that that's the direction I'd take. If you can do something like FindGeometricTrransform to get your two "perspectives" roughly aligned, you might be able to do something like explicitly compute image differences and find the region over which they're maximized. This is kind of on the cusp of "better with traditional methods" vs "better with machine learning", to my (admittedly somewhat ignorant!) mind.

My mistake, the 2-gram generated 'wore' and 'hi' and the 3-gram generated 'was'. Though I imagine the increase in the proportion of valid words in the sample is related to the distribution of word lengths in English?

Arden! I'm the first to plant my flag! Yay!!!!

I got the dark them working! Look, Mitch didn't register, lol :D

Attachments:

Screenshot - 04_...jpg

Oh hey, you mean Arten, ok, np!

Just kidding, my bad for the typo Arben, pls don't hurt me :D

Your code worked!

Attachments:

Screenshot - 05_...jpg