HN user

guard0g

589 karma

Work hard. Play harder. Help others.

Posts52
Comments92
View on HN
arxiv.org 21d ago

Reinforcement Learning with Metacognitive Feedback

guard0g
1pts1
arxiv.org 3y ago

Social Impacts of Reinforcement Learning with Human Feedback

guard0g
2pts0
bdtechtalks.com 3y ago

Large language models have a reasoning problem

guard0g
2pts1
arxiv.org 3y ago

On the Paradox of Learning to Reason from Data

guard0g
3pts0
vanmieghem.io 4y ago

Blueprint for evading endpoint protection in 2022

guard0g
1pts0
johncliu.github.io 4y ago

Taiwan Civil Defense Manual (English Translation) [pdf]

guard0g
12pts0
www.mnd.gov.tw 4y ago

Taiwan Military Issues Civil Defense Handbook [pdf]

guard0g
16pts5
www.reuters.com 4y ago

U.S. aircraft carrier deploys off Korean peninsula

guard0g
1pts0
www.france24.com 4y ago

Taiwan to move away from zero-Covid strategy

guard0g
3pts0
www.businessinsider.com 4y ago

Germany has seized control of Gazprom Germania

guard0g
277pts402
www.thelondoneconomic.com 4y ago

Chickens should wear masks to combat bird flu

guard0g
2pts0
www.reuters.com 4y ago

U.S. warns India, others against sharp rise in Russian oil imports

guard0g
3pts0
time.com 4y ago

Why Wearing a Face Mask Is Encouraged in Asia, but Shunned in the U.S. (2020)

guard0g
1pts0
www.nytimes.com 4y ago

What's the Deal with Masks on Planes?

guard0g
3pts2
uscnpm.org 4y ago

Possible Outcomes of the Russo-Ukrainian War and China’s Choice

guard0g
58pts34
www.forbes.com 4y ago

Shocking Statistics About College Graduation Rates

guard0g
2pts0
www.datascienceblog.net 4y ago

Effect Sizes: Why Significance Alone Is Not Enough (2018)

guard0g
1pts0
en.wikipedia.org 4y ago

Must read for many of our current events and societal debate

guard0g
2pts0
www.robotworkshop.com 4y ago

Hero – HeathKit's Robot from the 1980's

guard0g
2pts0
news.ycombinator.com 5y ago

Ask HN: Blog post about Stanford bad for undergrad entrepreneurs?

guard0g
1pts0
arxiv.org 6y ago

Weight Friction for Neural Networks

guard0g
1pts0
lifehacker.com 7y ago

Never use an airplane blanket again

guard0g
2pts0
www.forbes.com 7y ago

The Troubling Trajectory of Technological Singularity

guard0g
2pts0
github.com 7y ago

Project Alias: DIY Alexa/Google Home Privacy

guard0g
2pts0
www.bloomberg.com 7y ago

Quants Aren’t Adding Up

guard0g
2pts0
www.arvindguptatoys.com 7y ago

Parable by James Thurber (1939) as Relevant Today as It was Back Then [pdf]

guard0g
1pts1
www.nytimes.com 7y ago

Wielding Rocks and Knives, Arizonans Attack Self-Driving Cars

guard0g
191pts291
brohrer.github.io 7y ago

How to choose a machine learning model

guard0g
2pts0
ruder.io 7y ago

A survey of cross-lingual word embedding models (2016)

guard0g
4pts0
www.kdnuggets.com 7y ago

Better Transfer Learning for NLP

guard0g
2pts0

If I could relearn mathematics starting from elementary school, I'd get my hands on the AOPS books (Art of Problem Solving) and go through them on my own. They inspired my daughter sufficiently to pursue a math major in college.

Monitors: 1 - slack, teams, discord 2 - email, zoom 3 - markets monitor 4,5,6 - streaming news, tv 7, 8 - coding/work

I have the monitors mounted, stacked 2x2 on each side and tilted forming a semi-circle.

A lot of truth in this: '...young founders are more financially constrained than more experienced ones, leading them to cede upside to investors at a lower price. In other words, younger entrepreneurs may be a better “deal” for investors than more experienced founders.'

Colab is a cheap (free) way to start, though you won't be training very large models for very long (which you shouldn't be doing if you are a beginner). You learn a different set of skills when you put together your own rig and install/maintain the libraries, which is something I recommend everyone to try just go gain an appreciation of devops skills. But that's not a necessary diversion for a beginner (and may needlessly increase the learning curve).

It's not entirely painless for the West, given Russia's key position as supplier of wheat, fertilizer, and natural gas. Inflation will be kicking around for a while (even without trillion dollar stimulus programs) and creating economic distress for most.

This 996 culture also exists in Taiwan (a democracy). Hard-work is a core part of Chinese Confucian philosophy (as is competitiveness). Many of my Chinese & Taiwanese coworkers have told me they emigrated overseas to US and EU just so they can work less (while still working more than everyone else in the company).

While slavery is forced labor, 996 is more about people trying to get ahead. During my wall street days, I elected to work 8 am - 2 am, 7 days a week for subpar pay, all because of future potential rewards. Many of my peers dropped out due to the pressure and didn't stick around for the big payoffs. If you interview them, no doubt they would consider themselves "slave labor."

Labor costs are probably the major reason why US chip firms went fabless in the past few decades. Chip production in the lower 48 makes rational sense for national security but not necessarily economic sense. The higher labor costs will be passed through to the supply chain ultimately to spendthrift consumers.

Note the comparison to Wall Street investment banking culture, where you work 14+ hrs a day 7 days a week (and you wanted to).

When I was young and desperate for $ to survive (I couldn't even afford a can of Pepsi), I found that I could force myself to learn something I had no interest in or hated but was useful to where I wanted to be. Perhaps you just need the right perspective (desperation works!)

Get a full time job, make friends and treat everyone with respect, work hard and do a kickass job in something essential for the company (that other companies also need), then quit and offer to consult for them. They’re your first client, which builds credibility (and eliminates suspicion you left on less than optimal circumstances). Write a blog and/or volunteer in the local tech community, and you’ll be surprised how many referrals you can get. Also, other ex-employees of the company can be wonderful referrals. Good luck.

Too much effort on racial equality in the admissions process, too little on ensuring success in college, graduation and beyond. My top 10 college was hyper focused on diversity admissions, but the graduation rate of the under represented population was under 50%. Students shifted down due one tier due have a much higher rate of success than students shifted up one tier.

If we want to be serious about equity and equal opportunity, we should focus on eliminating legacy.