Going beyond “Sorry, I didn’t get that” building AI assistants that scale...
summary
Workshop shows how to build a machine‑learning assistant with Rasa NLU and Rasa Core, using Jupyter/Colab, CRF for NER, an SVM-style intent classifier and a TensorFlow embedding model for multi‑intent ranking. It trains on Rasa story format, wires a custom action that calls a paper‑search API, runs an action server, and enables interactive learning that updates the model live. Demo finds paper titles and authors despite tiny training data; production needs hundreds of examples per intent.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 2:26AI struggles with inputs
- 8:51Rasa open-source benefits
- 16:21Natural language understanding
- 23:10Gathering conversational data
- 33:34Training NLU model
- 51:57Dialogue management
- 1:00:16Domain and actions
- 1:13:54Interactive learning feedback
- 1:18:31Conversational AI challenges
- 1:21:54Q&A
related talks
01h 29mGoing Beyond “Sorry, I Didn’t Get That”: Building AI Assistants that Scale...
01h 38mConversational AI with Rasa Core & NLU
30mConversational AI: Building clever chatbots
01h 27mDeprecating the state machine: building conversational AI with the Rasa stack
24mConversational Interfaces for Chatbots and Artificial Intelligence
31mRasa: open source conversational AI to build next generation chatbots
32m"How Smart is AI? – Real projects to build real understanding"
46mFullstack AI: The Building Blocks for Production Ready AI-Powered Software