Tuesday Keynote
summary
Uses Python to gather and analyse academic metadata, builds coauthorship networks with NetworkX, and measures influence with centrality scores; reproduces Axelrod-style prisoner’s dilemma tournaments where tit‑for‑tat wins a 1,313‑entry tournament and the library contains 235 implemented strategies. Explains analytic solutions for memory‑one Markov policies via symbolic math, trains agents with recurrent nets and reinforcement learning using evolutionary‑generated training sequences, and models rhino dehorning with parameter sweeps to show when selective poaching becomes evolutionarily stable.
AI-written summary · about
chapters (11)
- 0:00Start
- 1:06Who Am I
- 4:22What Is a Pendulum
- 5:22The Double Pendulum
- 10:43Betweenness Centrality
- 13:20Vampire Bat
- 15:50Tit for Tat
- 19:17The Markov Chains
- 23:03Finite State Machine
- 30:50Why Does Python Have a Crash
- 34:45How Are You Selecting the Questions That You Are Going To Be Studying
related talks
39mKeynote 3: How did you know? Explaining Black Box Model Predictions in Python
18mKeynote: AI for Financial Services
49mIntroduction and Research Fellows Lightning Talks
37mSave the Princess
45mKeynote: Panel(...)
16mKeynote: Marin: An Open Lab for Frontier AI
43mGraphs: Datastructures to Query
02h 47mNetwork Analysis Made Simple