RL Hands-on Workshop - Train a Chatbot with Reinforcement Learning
#CellStratAILab #disrupt4.0 #WeCreateAISuperstars #PAAS #AIAsAService #MLOps #RL #API #Chatbots
Reinforcement Learning (RL) is an emerging area of Artificial Intelligence where we train agents based on rewards maximisation in state-action scenarios.
One very interesting application of RL is to use it to train Chatbots or Conversational bots.
A chatbot can be defined as a computer program that uses Natural Language to communicate with other humans or computers in the form of a dialogue. These are represented as Encoder-Decoder architectures in NLP. The Decoder outputs the probability distribution at every step (of the words to be predicted) - this is similar to RL policy gradient models. From this perspective, the decoder is seen as an agent trying to decide which token (RL Action) to produce at every step (RL state).
Hence a Chatbot Decoder can be tuned with help of RL algorithms such as Reinforce and Vanilla Policy Gradient (VPG).
Attend this Hands-on Workshop TOMORROW (Sat 12th Mar ’22) to learn about “Applying RL to train Chatbots” — this workshop will be presented on CellStrat Hub — one of the world’s most popular AI platforms.
Topic — RL Hands-on Workshop — Train a Chatbot with Reinforcement Learning
Date — TOMORROW (Sat 12th Mar ‘22), 10:30 AM IST
Presenter — Shubha M., AI Researcher and RL Expert
RSVP HERE TO ATTEND
See you in this Workshop tomorrow ! Let’s disrupt the world with AI, together !
ABOUT CELLSTRAT HUB :- Your one stop shop AI development and AI Apps platform. More than 13000 developers are already using this platform for AI. You too can get started here ! It comes bundled with 150+ projects in AI and ML domains, as well as numerous AI Apps for text processing and image processing tasks !
USEFUL LINKS :-
- CellStrat Hub
- CellStrat Hub AI Project Packs
- CellStrat Hub AI Apps and API Marketplace
- CellStrat YouTube Channel
- CellStrat product demo : Vision Apps and NLP Apps
- CellStrat AI Lab community (Disrupt4.0)
- CellStrat research : https://bit.ly/CS-research
