MIT 6.S094: Deep Reinforcement Learning

Показать описание

This is lecture 3 of course 6.S094: Deep Learning for Self-Driving Cars (2018 version). This class is free and open to everyone. It is an introduction to the practice of deep learning through the applied theme of building a self-driving car.

OUTLINE:
0:00 - AI Pipeline from Sensors to Action
8:25 - Reinforcement Learning
23:50 - Deep Reinforcement Learning
36:00 - AlphaGo
41:50 - DeepTraffic
54:35 - Conclusion

INFO:

CONNECT:
- If you enjoyed this video, please subscribe to this channel.

LINKS:

2017:

Рекомендации по теме

Комментарии

Can't thank you enough Lex, been following you since last year. Thanks again for all your hard work and keep sharing

rajat

One of the best lectures on application oriented deep learning! Top notch mind and speaker! Warmest thanks to you, Lex!

dennishuang

Thumbs up if you moved the cursor thinking it was yours but was Lex's one.

damiaFuentesFilms

can‘t tell if he‘s very bored or about to star in the next vin diesel movie

oleholgerson

This breaks down RL for me better than any other videos I have seen (and I have tried quite a few). Thank you very much for making this, and making it available publicly.

wamkong

What a wonderful video. Thanks so much for your effort, Lex! Given such open minded and extremely talented researchers, MIT remains on the top.

dennishuang

Clear overview of an important next step in AI, to appropriately extend end-to-end learning to reasoning, planning and acting, in a Deep Learning style (from 6:54 to 8:00). I would include transfer learning at different hierarchical levels as another major challenge, currently this technique is only applied at low hierarchical level, such as the first convolutional layers in CV or word embedding in NLP.

oswaldoludwig

from wikipedia: sampling is the selection of a subset (a statistical sample) of individuals from within a statistical population to estimate characteristics of the whole population

migueldias

Really excited, these videos are much more approachable than I had feared. I'm looking forward to the AGI videos coming out. Thanks for sharing.

DougPedley

I'm so jealous young generation. this high quality education is for free in internet. my girl friend want to marry but I need time to digest and build my own something. it should be there in my 20's. shit

kimchi_taco

Hey Lex I am a two time german go champion ( alas : very weak by global standard ^^ ) and i would love for you to explore that game and the intricacies that make it so interesting with me. Also to speak about the general questions of how to become good at something, the psychological, and physical aspects that fascilitate improvement.
Hope to speak to you soon
Best Regards
Jonas Welticke

iryumika

it's a pity that this lecture looks like a TEDx conference

hervediedie

6:12 This is why Skynet wants to extract revenge on humans.

brianhack

A great great lecture about reinforcement Learning 😍😍😍😍

RoxanaNoe

Opportunity to take class at MIT whose receiving strikes?

t.tucker

keep sharing ..thnku so much for these videos...much appreciated :)

reyanroy

The cats - they're Skinnerian cats, not Pavlovian ;)

pawebakalarz

To run reinforcement learning or Gan codes on windows can we use same level of gpu as we use for deep learning or we have different requirements???

ankitvaish

What computer language you use to program AI (in this case).

zbigniew

I really like Lex as a podcaster, but this is a just a list of buzz words put next to each other for an hour that make zero sense. Funny to watch though 😂

barnierguillaume

MIT 6.S094: Deep Reinforcement Learning

MIT 6.S094: Deep Reinforcement Learning for Motion Planning

MIT 6.S094: Deep Reinforcement Learning

MIT 6.S094: Introduction to Deep Learning and Self-Driving Cars

MIT 6.S094: Deep Learning

MIT 6.S091: Introduction to Deep Reinforcement Learning (Deep RL)

MIT 6.S094: Recurrent Neural Networks for Steering Through Time

MIT 6.S094: Deep Learning for Human-Centered Semi-Autonomous Vehicles

MIT 6.S094: Convolutional Neural Networks for End-to-End Learning of the Driving Task

MIT 6.S094: Deep Learning for Human Sensing

MIT 6.S094: Computer Vision

MIT 6.S191 (2023): Reinforcement Learning

Deep Learning Basics: Introduction and Overview

DeepTraffic Game (6.S094: Deep Learning for Self-Driving Cars)

DeepTraffic Solution | MIT: Deep Learning for Self-Driving Cars

MIT 6.S191 (2020): Reinforcement Learning

Lex Fridman: Is Reinforcement Learning a Good Candidate for AGI?

Deep Reinforcement Learning Nanodegree Program

MIT 6.S191 (2019): Deep Reinforcement Learning

Overview of Deep Reinforcement Learning Methods

MIT 6.S191 (2022): Reinforcement Learning

16. Reinforcement Learning, Part 1

MIT DeepTraffic 2.0 competition solution

Learning to perch a UAV on the ground using deep reinforcement learning

CS 285: Lecture 1, Part 1