Safety Gym

OpenAI Blog2019-11-21 16:000 views

We’re releasing Safety Gym, a suite of environments and tools for measuring progress towards reinforcement learning agents that respect safety constraints while training.

Source：https://openai.com/index/safety-gym

My favorites

PreviousBenchmarking safe exploration in deep reinforcement learning NextGPT-2: 1.5B release

Paradox of choice in design systems.

A good design system shouldn't maximize choices. It should minimize unnecessary decisions. In his famous TED Talk and bo

0dev.to

Global coordination requirements for 2038-class rollover events

Technical Paper ITU-T XSTP.epoch consolidates global understanding of risks associated with a cluster of structurally di

0Lobsters

Gentler Streak 6.0 adds a new wellbeing timeline and customizable dashboard

In an update rolling out today, Gentler Streak is introducing new ways for users to track their wellbeing and personaliz

09to5Mac

Lipinski Rule of Five Explained — With Worked Examples

The rule of five was published by Christopher Lipinski and colleagues in 1997, based on an analysis of compounds that ma

0dev.to

My ChatGPT plugin was approved. Then one real conversation found five problems

My plugin got approved for ChatGPT this week. It lets you describe something, see a print proof and a price, and get a r

0dev.to

I built an iOS app that turns forgotten screenshots into useful actions — Activae is finally live! 🚀

Hey everyone! 👋 I'm an indie iOS developer, and after months of development, testing, and plenty of debugging, I've fina

0dev.to

Safety Gym

Related posts

Paradox of choice in design systems.

Global coordination requirements for 2038-class rollover events

Gentler Streak 6.0 adds a new wellbeing timeline and customizable dashboard

Lipinski Rule of Five Explained — With Worked Examples

My ChatGPT plugin was approved. Then one real conversation found five problems

I built an iOS app that turns forgotten screenshots into useful actions — Activae is finally live! 🚀