bitwise-evan

Roughly 7 years ago I made a Sine Rider Solver: http://theartbot.com/labs/SineRiderSolver.html No idea if it still works.

According to my page, you called it "Super Cool".

> an adversary could trick Apple’s algorithm into erroneously matching an existing image

This is a very real, possible attack. Apple ships its CSAM model on device so any attacker can have a copy of the model. Then the attacker creates an image that triggers CSAM but looks like a panda [1]. Now the attacker sends tons of triggering photos to the unsuspecting victim, who now gets questioned by the FBI.

1: https://medium.com/@ml.at.berkeley/tricking-neural-networks-...

You are not legally allowed to sell or install used cats in CA: https://ww2.arb.ca.gov/aftermarket-catalytic-converters

I wonder if shops will still install used (stolen) units regardless? Or maybe they are just being shipped out of state or broken down for raw materials.

Are you asking what is the advantage of having ML optimized hardware on device? Yes, running inferences is expensive too, especially for video, photo, and speech processing. I would expect this phone to have user noticeable improvements in those three areas.

> The average cost of housing has gone up over time, but this trend is mainly driven by the fact that modern houses are larger and built better than houses in the past.

Oh, that explains why the house that sold for $300k in 1990 sold today for $2M (only 3X adjusted for inflation). It must have gotten 3X larger and 3X better built, even though it is the exact same house with a couple extra layers of paint and a redone kitchen.

Nice! I did something like this a long time ago: http://theartbot.com I wrote a couple algorithms but the best one basically treated the darkness of the image like the altitude in a map, then drew contour lines. I also physically built the plotter.

Likely because Google Assistant did not know it was running on a tv. It's probably the generic assistant that comes with Android. Not sure I would consider that pathetic. Tuning all the intricacies of the Assistant for each application seems like a ton of work.

I love these projects. I did a related project where I created a single, solid object that has 4 different silhouettes/shadows from 4 different angles in the same plane. Using more view angles than dimensions makes the problem much more complex. I suspect it is not possible to do this purely programmatically so I did it by hand had to smudge the letters to get everything to work out.

https://i.imgur.com/n1btEHG.mp4

https://www.thingiverse.com/thing:2111419

It is much easier to find security vulnerabilities if you have the source code.

Thank you so much for the kind words. I am the creator and love these kinds of projects. Check out another project I created along the same lines: TheArtBot.com

If you are asking why not many people do these types of projects- you sounds like an excellent candidate to do one yourself! I have list of project ideas on my phone that I will add to at random times. Next time you think of a cool idea write it down and make it a reality!

Yes- just like you are saying, the intention with the Mechancial Laser Show was not to maintain a constant laser speed, but accelerate and decelerate the laser gently. You can see the mechanism by which that is done by pausing the video at https://www.youtube.com/watch?v=_dtBUiaAqRE

Since the time taken to travel to each next point is constant, closer points together mean slower speed and spread out points mean high speed. You can see that the laser will slow down for taking the corners and speed up for the straights. There was no math behind this- I just eyeballed it but you could write software/math to help you determine the perfect spacing/acceleration.

Hi. This is Evan, the creator of Mechanical Laser Show. I am planning to present it at the upcoming Maker Faire in SF for a day so feel free to come by and check it out.

If you have any questions, feel free to ask here.