acadien

I agree with you there's a lot of anti SF content floating around these days. Just wanted to point out that in this article the author does go on to say he loves SF and thinks any uptick in crime will inevitably be countered. In that respect I felt it was a little different from the rhetoric you mentioned.

Does it go something like this?

Content creator posts content on their site that they own and pay to operate. Mega-3rd-party shares links of this content, sometimes posting blurbs from this content, driving traffic to Mega-3rd-party site and (to a much lesser extent) boosting traffic to content creator's site. However, Mega-3rd-party site also derives revenue from this by extracting information from user behavior and user interests, enabling them to improve ad-sales.

Am I misunderstanding something? Why is everyone here so pro mega-3rd-party?

Another question is this similar to remixing in the music industry? Feels like royalties in a twisted kind of way.

Saddles are a way of conceptualizing high dimensional optimization problems. If you have a 3 dimensional surface you can imagine a saddle as an isocurve that follows a minima in at least one dimension.

Another way to conceptualize these is to think of being at the minima of a parabola in 2 dimensions, but then seeing you're not in a minima in a 3rd dimension. Any time you're in a minima in at least 1 dimension, you're on a saddle.

You can extend this concept to a neural net which lives in millions of dimensions, undergoing SGD. When beginning an optimization run SGD moves in some direction to minimize the a bundled cost, inevitably stumbling into minima in (usually) many dimensions. Subsequent iterations will shift some dimensions out of minima and other dimensions into minima, the net is always living on a saddle during this process.

There are many papers that discuss the process in these terms and others that implicitly use it. I wouldn't say its a "hot area of research" but more of a tool for thinking about these processes and sometimes gaining some insight in to why things get stuck during training.

Hey @moultano in response to your argument about walls and Nets not being in a minima, its my understanding nets always live on high dimensional saddle points and that's commonly referred to in literature. Even when you're optimizing you're just moving towards ever lower cost saddles that are closer to the optimum but almost never a local optimum (for the reasons spelled out in your post).

I appreciate your response but respectfully disagree. (1) you posted above completely miss-states the facts. From the voter pamphlet text itself;

"A YES vote on this measure means: App-based rideshare and delivery companies could hire drivers as independent contractors. Drivers could decide when, where, and how much to work but would not get standard benefits and protections that businesses must provide employees."

If you drive full time you deserve full time benefits. That means more than a 50% payment of ACA (which is downright offensive to suggest someone can live off of while making minimum wage).

I am so against prop22 it pains me so here I am ranting on the internet.

What's going on is a negotiation between the state and these companies that are able to take advantage of a gap in how insurance and other benefits work. Instead of actively negotiating with the state, Uber/Lyft/Doordash and many others have simply asked to maintain the status quo, screw the drivers, just business as usual.

Drivers do not get benefits. They deserve benefits. Lets fix that.

Instead Uber/Lyft and others have chosen absolutely insane negotiating tactics. Literally their only solution is to

1. exempt all drivers from getting basic benefits like health insurance

or else

2. we're turning off our apps in California

This is madness. No on 22.

Depends on the application. Often down-sampled data is useful for drawing trends but not so useful for better understanding failure events.

This is a common misperception. You cannot estimate the shape of simple things in low light. Even humans are susceptible to optical illusions and are effectively blind during certain times of day (try driving in the direction of the sun during sun rise/set). Lidar is not susceptible to any of these things. Similarly ML will never solve low light or direct light situations, that is a physical limitation of cameras.

Time to start burying the powerlines fellow Californians.

The article said nothing about selective dispersal. For all we know these drones are just fancy crop dusters.

Car owners in small towns posting in here like "finally that fake gig economy is dying"

City dwellers posting the opposite because they rely on the product to make their lives work.

Here I am, stuck in the middle with you.

Age has nothing to do with talent. Older devs tend to have more experience, go figure.