Practical AI Reinforcement Learning for search
Oct 26, 2020
Hamish Ogilvy, founder of Sajari and an expert in machine learning and search technology, shares insights on the revolutionary impact of reinforcement learning in search. He discusses algorithmic trade-offs in ranking results and the integration of advanced AI technologies into search functionality. The conversation also touches on the evolution of search from traditional methods to voice queries, the challenges of data privacy, and the ongoing quest for real-time accuracy. Get ready to rethink how we search in the digital age!
AI Snips
Chapters
Transcript
Episode notes
Sajari's Origin
- Hamish Ogilvy's interest in search began during his physics PhD due to difficulties finding obscure research.
- This sparked his interest in conceptual search, leading to Sajari.
Speed-Accuracy Tradeoff
- Speed significantly impacts search performance and revenue, as demonstrated by Amazon's experiment.
- Balancing speed and AI's complexity is crucial for effective search.
Search-Driven UX
- Users prioritize search over navigation, expecting immediate, relevant results.
- Effective search reduces the burden on website design and reveals user intent.