The Love Symposium

Data scientist Matthew Fisher and I met at the Less Online conference at Lighthaven in Berkeley, where I gave a presentation on the Snake Cult of Consciousness. Between sessions, we discovered we had both been deep in the literature at the intersection of machine learning and personality psychology. In fact, he brought up some of my PhD research before knowing I was the author! Mark one for the annals of felicitous moments while haunting the halls of the rationalists.

Now we’re organizing the Love Symposium at the same location. Like Less Online, we expect to attract curious, open-minded people interested in novel approaches to old problems. The focus will be on using AI and personality science to address the problem of matching compatible people, particularly for long-term partnership.

If you're in the Bay Area November 15-17, come join us at Lighthaven. The main events will be during these dates, with more informal discussions happening in the days before. I'll personally be in SF November 11-25, so reach out if you'd like to meet up or if there’s someone you think I should connect with while I’m there.

Links & References:

• Dataclysm by Christian Rudder

• The Big Five personality traits

• TV Tropes database

• Tim Urban’s “How to Pick Your Life Partner”

Key Topics & Timestamps:

(00:00-01:18) Introduction: How Andrew and Matthew met at the Less Online conference, bonding over their shared interest in machine learning and personality research

(01:18-04:03) Matthew's Background: Working on image synthesis, StyleGAN, and human preference modeling in embedding spaces

(04:03-10:00) Discussion of the Love Symposium: An upcoming event exploring the intersection of technology, relationships, and human personality

(10:00-20:00) Historical Changes in Relationships:

• Impact of the Industrial Revolution vs progressive politics

• Case study of Japan's declining birth rates

• The role of technology in changing relationship patterns

(20:00-30:00) OkCupid vs eHarmony Approaches:

• Christian Rudder (OkCupid): Data-driven, algorithmic matching

• Neil Clark Warren (eHarmony): Expert system based on therapy experience

• Discussion of the "high variance" principle in dating profiles

(30:00-40:00) The Problem of Dating Apps:

• Network effects and startup challenges

• The potential of personality embedding

• Comparison to Netflix's recommendation system challenges

(40:00-50:45) TV Tropes Research:

• Using TV/movie character relationships to understand real relationship dynamics

• Building personality vectors and relationship prediction models

• Applications for dating apps and relationship matching

The event seeks researchers and startup founders interested in building next-generation social, friendship, and relationship-finding applications using AI and personality science.