Ciro Santilli @cirosantilli 37

The World Of Enrico Fermi by Harvard Project Physics (1970)

torchvision ResNet Updated 2025-07-16

pytorch.org/vision/0.13/models.html has a minimal runnable example adapted to python/pytorch/resnet_demo.py.

That example uses a ResNet pre-trained on the COCO dataset to do some inference, tested on Ubuntu 22.10:

cd python/pytorch
wget -O resnet_demo_in.jpg https://upload.wikimedia.org/wikipedia/commons/thumb/6/60/Rooster_portrait2.jpg/400px-Rooster_portrait2.jpg
./resnet_demo.py resnet_demo_in.jpg resnet_demo_out.jpg

This first downloads the model, which is currently 167 MB.

We know it is COCO because of the docs: pytorch.org/vision/0.13/models/generated/torchvision.models.detection.fasterrcnn_resnet50_fpn_v2.html which explains that

FasterRCNN_ResNet50_FPN_V2_Weights.DEFAULT

is an alias for:

FasterRCNN_ResNet50_FPN_V2_Weights.COCO_V1

The runtime is relatively slow on P51, about 4.7s.

After it finishes, the program prints the recognized classes:

['bird', 'banana']

so we get the expected bird, but also the more intriguing banana.

By looking at the output image with bounding boxes, we understand where the banana came from!

Figure 2.
python/pytorch/resnet_demo_out.jpg
. The beak was of course a banana, not a beak!

 Read the full article

Brightest natural objects in the sky Updated 2025-07-16

 View more

en.wikipedia.org/wiki/List_of_brightest_natural_objects_in_the_sky

 Read the full article

British Overseas Territories Updated 2025-07-16

 Read the full article

British physics research institute Updated 2025-07-16

 Read the full article

Tinker Tailor Soldier Spy (film) Updated 2025-07-16

 View more

This is not bad, but some divergences to the better BBC miniseries, which presumably sticks more closely to the novel:

in the film Jim Prideaux is captured in a cafe in Prague, in the series it's in the woods. It is therefore much more plausible that he would have been shot.
in the film Peter Guillam is played by Benedict Cumberbatch, who feels a bit young to be Ricki Tarr's boss. Not impossible, but still.
the series is much less chronological, and more flashback based, as new information becomes available. The film is more chronological, which makes it easier to understand, but less interesting at the same time.
in the film they shoot the Russian girl Irina in front of Jim, in the series the fact that she was shot is only known through other sources. The film has more eye candy, which weakens it.
Toby Esterhase is not threatened in an airfield, only in a safe ;house in London.

www.reddit.com/r/TrueFilm/comments/s5f3f1/tinker_tailor_soldier_spy_2011_is_the_best_spy/

 Read the full article

How large primes are found for RSA Updated 2025-07-16

 View more

Answers suggest hat you basically pick a random large odd number, and add 2 to it until your selected primality test passes.

The prime number theorem tells us that the probability that a number between 1 and

N

is a prime number is

1/ l o g (N)

Therefore, for an N-bit integer, we only have to run the test N times on average to find a prime.

Since say, A 512-bit integer is already humongous and sufficiently large, we would only need to search 512 times on average even for such sizes, and therefore the procedure scales well.

 Read the full article

Human Compatible Updated 2025-07-16

 View more

The key takeaway is that setting an explicit value function to an AGI entity is a good way to destroy the world due to poor AI alignment. We are more likely to not destroy by creating an AI whose goals is to "do want humans what it to do", but in a way that it does not know before hand what it is that humans want, and it has to learn from them. This approach appears to be known as reward modeling.

Some other cool ideas:

a big thing that is missing for AGI in the 2010's is some kind of more hierarchical representation of the continuous input data of the world, e.g.:
- intelligence is hierarchical
- we can group continuous things into higher objects, e.g. all these pixels I'm seeing in front of me are a computer. So I treat all of them as a single object in my mind.
game theory can be seen as part of artificial intelligence that deals with scenarios where multiple intelligent agents are involved
probability plays a crucial role in our everyday living, even though we don't think too much about it every explicitly. He gives a very good example of the cost/risk tradeoffs of planning to the airport to catch a plane. E.g.:
- should you leave 2 days in advance to be sure you'll get there?
- should you pay an armed escort to make sure you are not attacked in the way?
economy, and notably the study of the utility, is intrinsically linked to AI alignment