MIT 6.S191 (2018): Sequence Modeling with Neural Networks

Показать описание

MIT Introduction to Deep Learning 6.S191: Lecture 2
Sequence Modeling with Neural Networks
Lecturer: Harini Suresh
January 2018

Рекомендации по теме

Комментарии

This is exactly what I was looking for. Thank you

omkarpanhalkar

Excellent video, great mathematical explanation !

ahanmr

thank you very much, very good sequence

victorarias

2:10 okay the traditional feed-forward NN takes a fixed-length vector I get that part. In sequence modeling do we need to fix these issues and make fixed lengths as well? so that means both traditional feed-forward or sequence modeling architecture both need to be a fixed-length vector?

AjayTiwari-fnpz

omg, loved the part of the generated music!

lex

hi, @alexanderamini are there recording for the lab session. or I can learn them from the Jupiter notebooks good enough.

NikhilKumar-pzuz

is the first lecture of this course the only prerequisite for this lecture on sequence modeling?

abhishekravikumar

5:56 Very interesting problems! It seems to me that using a really big fixed window approach (i.e., max words in a sequence) is worth exploring further and should not be given up so quickly.

shakesbeer

i feel that the only reason im understanding the material is that i know of it in the first place.I think that if i was being introduced to it by these lectures so far i wouldnt have got a chance to get anything more than a general idea ..And i mean about the theory cause i really doubt the lab sessions would shed any more light to back propagation and back propagation through time to myltilayered rnns other than how to use them in action as black boxes.

bibiskga

Great lecturers and super clear slides. Are the labs you mention in the video available somewhere?

dariobonaretti

*where can i learn about the math necessary for ML in 2018/2019?!?!*

joebazooks

could anyone explain in simple words why parameter sharing is required

princeroshan

Yes, the lecture is very good and enlightening.
The girl, however, should better not give the
lecture. Her way of speaking and her pitch
and her voice are not pleasant to listen to.

silberlinie

MIT 6.S191 (2018): Sequence Modeling with Neural Networks

MIT 6.S191 (2018): Sequence Modeling with Neural Networks

MIT 6.S191 (2018): Deep Generative Modeling

MIT 6.S191 (2018): Deep Learning - A Personal Perspective

MIT 6.S191 (2018): Deep Learning Limitations and New Frontiers

MIT 6.S191 (2018): Introduction to Deep Learning

MIT 6.S191 Lecture 2: Sequence Modeling with Neural Networks

MIT 6.S191 (2018): Computer Vision Meets Social Networks

MIT 6.S191 (2018): Deep Reinforcement Learning

MIT 6.S191 (2018): Convolutional Neural Networks

MIT 6.S191 (2018): Issues in Image Classification

MIT 6.S191 (2018): Faster ML Development with TensorFlow

MIT 6.S191 (2021): Recurrent Neural Networks

MIT Introduction to Deep Learning (2022) | 6.S191

MIT Introduction to Deep Learning (2023) | 6.S191

MIT 6.S191 (2019): Recurrent Neural Networks

MIT 6.S191 (2020): Introduction to Deep Learning

MIT 6.S191 (2020): Recurrent Neural Networks

MIT 6.S191 (2018): Beyond Deep Learning: Learning+Reasoning

MIT 6.S191 (2019): Introduction to Deep Learning

MIT 6.S191 (2019): Deep Generative Modeling

MIT 6.S191 (2021): Deep Learning New Frontiers

MIT 6.S191: (Google) Generative AI for Media

MIT 6.S191 (2020): Deep Generative Modeling

Introduction to Deep Learning (MIT 6.S191)