0% found this document useful (0 votes)
10 views29 pages

Evolution of Language Models: RNN to BERT

The document discusses the evolution of language models from Recurrent Neural Networks (RNNs) to BERT, highlighting the limitations of RNNs in handling long-term dependencies and the advantages of Transformer models. It explains BERT's bidirectional approach to language understanding through techniques like Masked Language Model and Next Sentence Prediction. The document also emphasizes the importance of fine-tuning BERT for various downstream tasks.

Uploaded by

Sachin Kumar N
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
10 views29 pages

Evolution of Language Models: RNN to BERT

The document discusses the evolution of language models from Recurrent Neural Networks (RNNs) to BERT, highlighting the limitations of RNNs in handling long-term dependencies and the advantages of Transformer models. It explains BERT's bidirectional approach to language understanding through techniques like Masked Language Model and Next Sentence Prediction. The document also emphasizes the importance of fine-tuning BERT for various downstream tasks.

Uploaded by

Sachin Kumar N
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Language Models

From RNN to BERT


Neural LM
 Unlimited History: Recurrent Neural Networks
- The RNNs are neural nets that can deal with sequences of
variable length. They are able to this by defining a recurrence
relation over timesteps:

20 Credit: Speech and Language Processing. Daniel Jurafsky & James H. Martin.
Neural LM
 Unlimited History: Recurrent Neural Networks
- The activation value of the hidden layer depends on the
current input as well as the activation value of the hidden
layer from the previous time step.
- In this way, Recurrent neural language models (RNNLMs)
avoid the limited context constraint inherent in traditional N-
gram models, since the hidden state embodies information
about all of the preceding words all the way back to the
beginning of the sequence.

21 Credit: Speech and Language Processing. Daniel Jurafsky & James H. Martin.
Neural LM
 Unlimited History: Recurrent Neural Networks
- The activation value of the hidden layer depends on the
current input as well as the activation value of the hidden
layer from the previous time step:

22 Credit: Speech and Language Processing. Daniel Jurafsky & James H. Martin.
Neural LM
 Unlimited History: Recurrent Neural Networks

23 Credit: Speech and Language Processing. Daniel Jurafsky & James H. Martin.
Neural LM
 Unlimited History: Recurrent Neural Networks

24 Credit: Speech and Language Processing. Daniel Jurafsky & James H. Martin.
RNN variants
 Gated RNNs
- Credit assignment issue
- Vanishing gradient issue

25 Credit: Speech and Language Processing. Daniel Jurafsky & James H. Martin.
RNN Variants
 Gated RNNs
 Bi-directional RNNs

26 Credit: Speech and Language Processing. Daniel Jurafsky & James H. Martin.
RNN Features
 Gated RNNs
 Bi-directional RNNs

27 Credit: Speech and Language Processing. Daniel Jurafsky & James H. Martin.
Magic and Reality
 Magic: Flexible
- complex networks can be treated as modules that can
be combined in creative ways.
- Naturally suited for variable-length sequence processing
- Theoretically handling long-term dependencies
 Reality: hard to train
- Long-range dependencies still tricky, despite gating
- Limited contextual information are captured
- Loss of hierarchy
- Sequentiality prohibits parallelization within instances
- Training is slow

28
Transformers
 Attention is All You Need
- Transformer is a pure attention-based model consisting of:
- The encoding component is a stack of encoders.
- The decoding component is a stack of decoders of the
same number.
 Transformer Block
- The encoders are identical Transformer Block.
- The encoder architecture is also used in BERT.

39
Transformers
 Attention is All You Need
- Transformer is a pure attention-based model consisting of:
- The encoding component is a stack of encoders.
- The decoding component is a stack of decoders of the
same number.
 Transformer Block
- The encoders are identical Transformer Block.
- The encoder architecture is also used in BERT.

40
Transformer Block

41 Credit: The Illustrated Transformer. Jay Alammar.


Transformer Block

42 Credit: The Illustrated Transformer. Jay Alammar.


Transformer Block

43 Credit: The Illustrated Transformer. Jay Alammar.


Transformer Block
 Attention is All You Need
- Attention: indexing/referring contents using similarity
- e.g. Encoder-Decoder Attention

44
Transformer Block
 Attention is All You Need
- Attention: indexing/referring contents using similarity
- Self-Attention

45
Transformer Block
 Attention is All You Need
- Attention: indexing/referring contents using similarity
- Convolution v.s. Self-Attention

46
Transformer Block
 Attention is All You Need
- Attention: indexing/referring contents using similarity
- Self-Attention (similar with convolution)
- Constant path length between any two positions
- Variable-sized perceptive field
- Gating/multiplication enables crisp error propagation
- Trivial to parallelize (per layer)
- Long-distance context has “equal opportunity”
- Self-attention is the method the Transformer uses to bake
the “understanding” of other relevant words into the one
we’re currently processing.

47
BERT
 Bidirectional Encoder Representations from Transformers:
BERT
- BERT ≈ Encoder of Transformer
 Combining the Two Worlds:
- BERT ≈ ELMo + Transformer
- ELMo’s (Embeddings from Language Models) language
model was bi-directional, but based on (slow) RNNs
- Transformer purely relies on (parallelized) attention,
but only trains a forward language model.
- Why: Language models only use left context or right
context, but language understanding is bidirectional.

65
BERT
 Handling the bi-directionality:
- Transformer’s encoder is only a forward language model
- If they were to perform bidirectional self-attention, then the
model would learn that the next word in the sentence is the
target and would predict it always, with 100% accuracy.

66
BERT
 Handling the bi-directionality:
- Transformer’s encoder is only a forward language model
- If they were to perform bidirectional self-attention, then
the model would learn that the next word in the sentence is
the target and would predict it always, with 100%
accuracy.
- BERT solves this problem using two “new” paradigms:
- Masked Language Model
- Next Sentence Prediction (Two-Sentence Tasks)

67
Masked Language Model
 Predict random words from within the sequence
- Transformer’s encoder is only a
forward language model
- If they were to perform
bidirectional self-attention,
then the model would learn
that the next word in the
sentence is the target and
would predict it always, with
100% accuracy.
- BERT solves this problem using
two “new” paradigms:
- Masked Language Model
- Two-sentence Tasks
68 Credit: The Illustrated BERT. Jay Alammar.
Masked Language Model
 Predict random words from within the sequence
- In this specific case, 15% of the words that were fed in
as input were masked.
- Too little masking: Too expensive to train
- Too much masking: Not enough context

69 Credit: The Illustrated BERT. Jay Alammar.


Masked Language Model
 Predict random words from within the sequence
- In this specific case, 15% of the words that were fed in
as input were masked.
- But not all tokens were masked in the same way:
- 80% of the time, replace with [MASK]
- went to the store → went to the [MASK]
- 10% of the time, replace random word
- went to the store → went to the running
- 10% of the time, keep same
- went to the store → went to the store

70 Credit: The Illustrated BERT. Jay Alammar.


Masked Language Model
 Predict random words from within the sequence
- In this specific case, 15% of the words that were fed in
as input were masked.
- But not all tokens were masked in the same way since:
- If the model had been trained on only predicting
‘<MASK>’ tokens and then never saw this token during
fine-tuning, it would have thought that there was no need
to predict anything.
- The model would have only learned a contextual
representation of the ‘<MASK>’ token and this would
have made it learn slowly (since only 15% of the input
tokens are masked).

71 Credit: The Illustrated BERT. Jay Alammar.


Next Sentence Prediction
 To learn relationships between sentences
- predict whether Sentence B is actual sentence that
proceeds Sentence A, or a random one
- BERT ≈ Encoder of Transformer with Input Specifics
- [CLS]
- [SEP]

7 Credit: The Illustrated BERT. Jay Alammar.


Next Sentence Prediction
 Predict random words from within the sequence
- Transformer’s encoder is only a forward language model
- If they were to perform bidirectional self-attention, then the
model would learn that the next word in the sentence is the
target and would predict it always, with 100% accuracy.
- BERT solves this problem using two “new” paradigms:
- Masked Language Model
- Two-sentence Tasks

73 Credit: The Illustrated BERT. Jay Alammar.


7
Two-Stage Fine-Tuning
 Fine-tuning for downstream tasks
- The BERT paper shows a number of ways to use BERT for
different tasks.

74

You might also like