Start Here
Who I am
I'm a student working toward original research in neuroscience, recording how I think we actually get there.
Ponder
Unavailable… for now.
What I'm investigating
Neuroscience, AI Ethics, AGI — see Research for what I actually want to understand more deeply.
Current research
RNN Flip-Flop Study → — Regularized vs. unregularized RNNs on the 3-bit flip-flop task, analyzed with Sussillo & Barak's fixed-point method.
Latest writing
How I Define AGI → — A follow-up essay: a working definition of AGI, how it will come about, and the precautions it needs.
Something I built
RNN Flip-Flop Study → — Regularized vs. unregularized RNNs on the 3-bit flip-flop task, analyzed with Sussillo & Barak's fixed-point method.
What I'm trying to figure out
Does regularization make small RNNs more robust to noise, or did the regularized network just start from a lower baseline loss? Retraining both from different random starting points, with matched training, is the next test.