- RivieraKidBut how could they beat Waymo or Tesla? At some point, these companies will offer an L4 package to car makers.
- My first instinct is also that Rivian's strategy doesn't make sense. Self-driving is a monumentally hard problem, to be successful you need a world-class engineering and research team, resources and time.
I suspect that when Rivian has an L3 product, Waymo will be already offering an L4 package to car manufacturers.
- I've been saying that it's semi-solved, in the sense that we have a decent idea of how to get there without requiring major breakthroughs. (By "we" I mean Waymo.)
- > Google Maps is not just indexing demand - it is actively organising it through a ranking system built on
This is where I stopped reading.
- Yes, it's an early stage technology and the logistics of scaling is non-trivial. Yet, if you look at the numbers, they've been scaling surprisingly fast, at a sustained rate of 5x per year for 5 years in weekly paid rides.
- They've been scaling paid rides per week by about 5x every year for the past 5 years, that is fast in my book.
- Waymo have been driving on highways 15 years ago in supervised mode. And 1 or 2 years ago unsupervised.
- Driving at the speed limit will not meaningfully increase the chance of an accident.
- Why? The press release is much more useful for the vast majority of HN readers in my opinion. The paper is something you read if you want to know more so the right place for it is the comments.
In general, not referring to this specific case, scientific papers are often written for people with specialized background and are hard to understand for people without that background, even if they're otherwise smart and educated.
- What are you working on specifically? I've been vaguely following poker research since Libratus, the last paper I've read is ReBeL, has there been any meaningful progress after that?
I was thinking about developing a 5-max poker agent that can play decently (not superhumanly), but it still seems like a kind of uncharted territory, there's Pluribus but limited to fixed stacks, very complex and very computationally demanding to train and I think also during gameplay.
I don't see why a LLM can't learn to play a mixed strategy. A LLM outputs a distribution over all tokens, which is then randomly sampled from.
- The car can ask for remote assistance, which doesn't happen often.
- Waymo, which works and is scaling quickly.
- Waymo's self-driving cars are scaling quickly. With some inaccuracy it can be said that the problem is solved, we have the technology for a full-scale deployment, we just need to do the boring work to deploy it everywhere.
- This would be great if true, I need 5 years to reach financial independence, so a decade should be plenty of time.
- We're nowhere close to solving aging. We don't even understand aging and understanding the problem should be much easier than solving it.
- Can anyone link a Wikipedia article or briefly describe how does this work?
- Could this technology be modified to 1 to 1 video calls?
- But there was no car in front of the Waymo, it was a cat.
- I remember paper models being very widespread when I was a kid in the Czech Republic, they were always included in a popular magazine for kids, no idea whether it has changed. Per ChatGPT this is unique for this region - Czech Republic, Poland, Slovakia.
- By the time this becomes a problem, seniors won't be needed either.