explaingit

oxford-cs-deepnlp-2017/lectures

★ 15,855Audience · researcherComplexity · 1/5Setup · easy

TLDR

Lecture slides, video recordings, and coding exercises from Oxford University's 2017 deep learning for natural language processing course, co-taught with DeepMind researchers.

Mindmap

mindmap
  root((repo))
    Topics
      Word embeddings
      Recurrent networks
      Machine translation
      Speech recognition
    Resources
      PDF slide decks
      Video recordings
      Coding exercises
    Institutions
      Oxford University
      DeepMind
    Audience
      ML researchers
      Advanced students
Click or tap to explore — scroll the page freely

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

Things people build with this

USE CASE 1

Work through the full Oxford Deep NLP 2017 curriculum to build a foundation in neural networks for language tasks.

USE CASE 2

Use the practical coding exercises to get hands-on experience with word embeddings and recurrent neural networks.

USE CASE 3

Watch video lectures from DeepMind and CMU researchers on machine translation, speech recognition, and question answering.

Getting it running

Difficulty · easy Time to first run · 5min

In plain English

This repository contains the lecture slides and materials from Oxford University's Deep Natural Language Processing (NLP) course, delivered in early 2017 in partnership with researchers from DeepMind. NLP is the field of teaching computers to understand and generate human language, things like translating text, answering questions, or transcribing speech. The course is advanced and assumes prior knowledge of machine learning. It covers how neural networks (AI systems inspired loosely by the brain) can be applied to language tasks. Topics include word embeddings (turning words into numbers that capture meaning), recurrent neural networks (models that process sequences of words one at a time), and how these can be used for tasks like language translation, speech-to-text, and question answering. Each lecture has PDF slides and a video recording linked directly in the README. There are also practical coding exercises for hands-on experience with the concepts covered. The course was organized by Phil Blunsom and included lecturers from Carnegie Mellon University, DeepMind, and NVIDIA. The full README is longer than what was provided, but the repository is essentially an open-access archive of the course's educational materials.

Copy-paste prompts

Prompt 1
I'm working through the Oxford Deep NLP 2017 course on recurrent neural networks. Help me implement a basic RNN language model in Python based on the concepts from that curriculum.
Prompt 2
Based on the Oxford 2017 Deep NLP course, explain how word embeddings capture semantic meaning and help me write code to train simple word vectors on a small text corpus.
Prompt 3
Help me replicate the sequence-to-sequence translation architecture described in the Oxford Deep NLP 2017 lectures, starting from the encoder-decoder concept in the slides.
Prompt 4
Using the Oxford 2017 NLP course as a guide, help me build a simple question-answering system that takes a passage and a question as input and returns a span answer.
Open on GitHub → Explain another repo

← oxford-cs-deepnlp-2017 on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.