HNHacker News
TopNewBestAskShowJobs

young_rutabaga

47 karma · joined April 29, 2020

submissionscomments
young_rutabaga··on Bike Bureau: Report Bike Lane Obstructions
My favorite one is a "Share the Road" sign with a bike on it, intended to remind drivers of bikers' presence -- but placed in the middle of a bike lane. Palo Alto loves these for some reason
young_rutabaga··on eBay explicitly bans AI "buy for me" agents in user agreement update
Snipers essentially convert the ascending-bid proxy auction used in eBay into a Vickrey second-price sealed bid auction, allowing a buyer to not reveal their preferences to other participants. In theory, with rational participants, this shouldn't have any effect on revenue. In practice, buyers do not always understand auction mechanics and delay setting the highest price they're willing to pay until they are outbid. If they're outbid 3 seconds before the deadline, they lost.
young_rutabaga··on Nine US states are teaming up to accelerate the adoption of heat pumps
It's worth checking if room temperature is measured at the thermostat or at the return. It's a common issue with ducted Mitsubishi systems.

Also, a thermostat should be away from the air flow from supply registers to measure representative temperature

young_rutabaga··on Nine US states are teaming up to accelerate the adoption of heat pumps
Inverter heat pumps shouldn't be controlled by dumb "smart" thermostats like Nest. They send only on-off signals, while a compatible communicating thermostat sends a numerical setpoint, allowing the unit to modulate
young_rutabaga··on Thousands of Mazdas in the Seattle area are stuck on a single FM radio station
I had this issue, too. It helped to replace the AA headunit module in AIO with a binary having this fix https://github.com/gartnera/headunit/pull/174. Since then, AA works great for me
young_rutabaga··on Using Reinforcement Learning in the Algorithmic Trading Problem
High turnover, high costs. You flip your position too often
young_rutabaga··on Using Reinforcement Learning in the Algorithmic Trading Problem
This paper is a hilarious dump of WQ's randomly generated formulas that (hopefully) happen to pass in-sample test.

  Alpha#33: rank((-1 * ((1 - (open / close))^1)))
This formula trivially reduces to

  rank(open/close - 1)
which is an example of a mean-reversion strategy. But: 1) nobody bothered to simplify this formula, 2) as any mean reversion, it is extremely difficult to trade.