So you can discuss your weekly readings.
I’m still reading Moby Dic, currently halfway through, so far is great, though as English is not my native language It means I have to look up quite a bit in the dictionary to learn the “old timey” words, but I’m using it less now the more I learn, it also made me check up the Bible to find the biblical counterparts of the characters with biblical names, which is interesting.
.webp?h=8 "Merry Fistmas!")
Jump in the discussion.
No email address required.
Writing a paper on Hierarchical Reinforcement Learning in specific settings, so I need to do a literature survey on the topic.
Previously, I had limited myself to simplistic settings, so this time I thought I'll try something more interesting and started working on a robotic setting similar to that in Hierarchical Reinforcement Learning With Universal Policies for Multistep Robotic Manipulation.
The goal is to teach a robotic gripper to handle Lego-like blocks like a toddler. The approach in the paper mentioned above is to learn a goal-conditioned high-level policy that (implicitly) plans and creates low-level goals, that are then given to a low level policy that actually does the work to achieve these goals. Much like how I do my work, ironically.
I didn't fully get into the details of the algorithm, which is a combination of multiple techniques including DDPG, HER and more, let alone the actual code. But we have verified that the author's official code actually performs as advertised, so that's a start.
Jump in the discussion.
No email address required.
Don't care, didn't ask 😘 post this on ure blog instead next time sweaty
Jump in the discussion.
No email address required.
Jump in the discussion.
No email address required.
More options
Context
More options
Context
I took one (in numbers: 1) class during my Master's in Reinforcement Learning and noped out right after that.
God, you're really a fricking nerd.
Jump in the discussion.
No email address required.
You claim to be an incel, yet you noped out of Reinforcement Learning
Curious.
I was also afraid of getting into RL. I was doing research in generic ML and some niche theoretical AI related subfields
but my guide slowly manipulated me over the course of two years into changing my field, it took me by surprise lol.
Also, do people get masters in RL? It's such a specific subfield, wouldn't it be a masters in CS or AI?
Jump in the discussion.
No email address required.
I meant one class about RL during my degree in a CS adjacent field. For some reason my English keeps getting worse and worse, it's honestly very depressing.
That's the power of advisors and one more reason why academia is in the state it is currently in.
Jump in the discussion.
No email address required.
More options
Context
More options
Context
More options
Context
More options
Context