59:15
Can we solve inequality in South Africa? Interview with Dieter von Fintel (TGIF 2024)
14:41
Reinforcement learning from human feedback (NLP817 12.3)
13:10
The difference between GPT and ChatGPT (NLP817 12.2)
13:56
Large language model training and inference (NLP817 12.1)
07:41
Extensions of RNNs (NLP817 9.7)
10:18
Solutions to exploding and vanishing gradients (in RNNs) (NLP817 9.6)
12:59
Vanishing and exploding gradients in RNNs (NLP817 9.5)
24:41
Backpropagation through time (NLP817 9.4)
03:22
RNN definition and computational graph (NLP817 9.3)
08:31
RNN language model loss function (NLP817 9.2)
15:04
From feedforward to recurrent neural networks (NLP817 9.1)
10:19
Embedding layers in neural networks
16:24
Git workflow extras (including merge conflicts)
23:26
A Git workflow
21:11
Evaluating word embeddings (NLP817 7.12)
11:58
GloVe word embeddings (NLP817 7.11)
15:47
Skip-gram with negative sampling (NLP817 7.10)
05:35
Continuous bag-of-words (CBOW) (NLP817 7.9)
02:30
Skip-gram example (NLP817 7.8)
09:47
Skip-gram as a neural network (NLP817 7.7)
09:49
Skip-gram optimisation (NLP817 7.6)
07:32
Skip-gram model structure (NLP817 7.5)
07:59
Skip-gram loss function (NLP817 7.4)
07:28
Skip-gram introduction (NLP817 7.3)
06:19
One-hot word embeddings (NLP817 7.2)
08:57
Why word embeddings? (NLP817 7.1)
24:00
What can large spoken language models tell us about speech? (IndabaX South Africa 2023)
04:12
Hidden Markov models in practice (NLP817 5.13)
08:33
The log-sum-exp trick (NLP817 5.12)
12:15
Why expectation maximisation works (NLP817 5.11)
20:18
Soft expectation maximisation for HMMs (NLP817 5.10)
12:06
Hard expectation maximisation for HMMs (NLP817 5.9)
08:29
Learning in HMMs (NLP817 5.8)
19:12
The forward algorithm for HMMs (NLP817 5.7)
07:11
Why do we want the marginal probability in an HMM? (NLP817 5.6)
19:16
Viterbi HMM example (NLP817 5.5)
23:50
The Viterbi algorithm for HMMs (NLP817 5.4)
02:54
The three HMM problems (NLP817 5.3)
08:35
Hidden Markov model definition (NLP817 5.2)
13:52
A first hidden Markov model example (NLP817 5.1)
14:20
What are perplexity and entropy? (NLP817 4)
02:07
Are N-gram language models still used today? (NLP817 3.12)
07:51
Kneser-Ney smoothing (NLP817 3.11)
04:28
Language model backoff (NLP817 3.10)
10:36
Language model interpolation (NLP817 3.9)
04:37
Absolute discounting in language models (NLP817 3.8)
10:26
Additive smoothing in language models (NLP817 3.7)
06:00
Language model smoothing intuition (NLP817 3.6)
08:47
Evaluating language models using perplexity (NLP817 3.5)
03:46
Why use log in language models? (NLP817 3.4)
10:47
Start and end of sentence tokens in language models (NLP817 3.3)
12:57
N-gram language models (NLP817 3.2)
10:08
The language modelling problem (NLP817 3.1)
03:35
Transformer (NLP817 11.10)
07:27
Cross-attention (NLP817 11.9)
04:43
Masking the future in self-attention (NLP817 11.8)
05:22
Multi-head attention (NLP817 11.7)
05:02
The clock analogy for positional encodings (NLP817 11.6)
19:29
Positional encodings in transformers (NLP817 11.5)
04:44
Self-attention in matrix form (NLP817 11.4)
X
About
Tags