Professional Documents
Culture Documents
Gmail Smart Compose
Gmail Smart Compose
Gmail Smart Compose
Beam Search
02 Smart Reply
05 Algorithm
➢ Scale - 1.5 billion diverse users. The model needs to have enough capacity so that
it is able to make tailored, high-quality suggestions in subtly different contexts.
➢ Personalisation - Users often have their own unique email writing styles
➢ Fairness and Privacy - Needed to address sources of potential bias in the training
process,
Fairness and Privacy
Date & Time - For suggesting “Good Morning” , “Happy New Year” ..etc
Definition - Model to find the probability of the next word or sequence of words.
Yt = f (Yt-1 )
Objective - To maximise the log probability of producing the correct target sequence.
hi - Hidden Layer
yi - Input
Wi - Weight
xi - output
Sequence-to-Sequence model
3
BoW(Bag of Words) Model
It was the best of times,
it was the worst of times,
it was the age of wisdom,
it was the age of foolishness
BoW RNN
2. If n is the goal state, backtrace path to n (through recorded parents) and return path.
5. If |OPEN| > ß , take the best ß nodes (according to heuristic) and remove the others from the OPEN.
done
Train the
model
Personalisation
During the prefix-encoding phase, the goal is to obtain the final hidden states
of the language model, which represent the full encoding of the prefix. The
beam-search phase then explores hypothetical extensions of the prefix,
keeping track of plausible candidate suggestions.
1. Gmail Smart Compose: 1. https://arxiv.org/pdf/1906.0
Real-Time Assisted 0080.pdf
Writing
1. https://ai.googleblog.com/2
1. Google AI Blog 018/05/smart-compose-
using-neural-networks-
to.html