officehero

9 ms·

61.
by officehero 6y ago
Your post is proof that HN is far more than 4000 ppl. You represent the average consumer.
62.
by officehero 6y ago
The educated guess for the last 50 years was 20-30 years away
63.
by officehero 6y ago
> become boring 80-95% of "ML" has always been "boring" and any data scientist/ML-engineer worth their salt know this. This also concerns what you refer to as "pure ML job". It only takes fresh grads an
64.
by officehero 6y ago
I also think FB, contrary to many other massive enterprises, suffers extra much from this bcs it doesn't even make sense as a massive enterprise. It's a social network platform and when it grew too large it mutated into something
65.
by officehero 6y ago
The main problem is not backpropagation though, but the fixation of resources on DL projects (that's what I call local minimum!). In my department, for example, they don't seem to care about the application, integration, deploymen
66.
by officehero 6y ago
Is this related to the "build things that don't scale" motto? That one was conceived for startups in general if I remember correct. Is this same thing but in more detail?
67.
by officehero 6y ago
> Taking on the responsibility of pushing back hard on poorly conceived new features is one of the important hidden skills of being an effective software developer. example from my current project: 1. Inherit half finished large software
68.
by officehero 6y ago
The big problem is that the populace do not understand statistics. As soon as a self driving car kills someone the witch hunt begins and the whole fleet needs to be retracted until investigation is over. Can your image network solve this pr
69.
by officehero 6y ago
There is a push to upgrade road infrastructure with sensors so might be sooner. You're totally right about regulation though - engineers are only solving a "technical" problem, but what about the non-technical ones? Where is
70.
by officehero 6y ago
The "key" is a total abuse of the term "AI". The dictionary definitions basically say "machines that appear intelligent", and "intelligence" is "ability to acquire knowledge [/skills throug
71.
by officehero 6y ago
Interesting point of view. So as long as a paper is "non-tabloid" it should be more trusted. A Chinese or Russian state owned non-tabloid is then more trustworthy than NYP? Propaganda is everywhere and the definition of what and w
72.
by officehero 6y ago
Although the "data problem" is already well known by all ML and RL engineers/researchers (with varying degree depending on what one works on), it's an innovative and concise post that really boils it down. It's part
73.
by officehero 6y ago
Extending these types of examples to 100 doors with 99 goats of which Monty opens 98 makes it even easier.
74.
by officehero 6y ago
Disagree on the "playground" historical note students hundreds of years ago would probably not agree at all, basically since life back then was tougher for everyone including elite compared to tosay. Somewhat agree with 2d and 3d
75.
by officehero 6y ago
Doesn't this just mean that the "reward" definition is poor in this case i.e. the agent does not care about maximizing it because there are additional components left out of it (e.g. survival)?
76.
by officehero 7y ago
SOYLENT, SOYLENT, SOYLENT, SOYLENT ...
77.
by officehero 7y ago
The correct answer in this case is of course YES, it's justified. It's a hedge fund after all. Motivate with evolutionary theory starting with Darwin and ending with Nietzsche, maybe even bring in Sun Tzu depending on interviewer.