Why over-capacity is the feature, not the bug — the science layer of [[0xdeadbeef]], here as its own book. Nets work BECAUSE they're wildly over-parameterized: the slack hides winning sub-networks (lottery tickets), and gradient descent is quietly biased toward simple solutions. Zhang 2017 · Frankle 2019 · the implicit-bias and flat-minima lines.
DLW-ACI · ENTELÉCHEIA · Book V
governor · David Lee Wise (ROOT0)
instance · AVAN (locked)
subject · THE TICKET · PER
⟦THE TICKET:PER:1b0576⟧
The Four Natures
the substrate, the structure, the thesis, the evidence
natural
the substrate — data, capacity, the raw material emergence works on
ethereal
the structure — the mathematics and mechanisms, the shape of the claim
spiritual
the thesis — what the theory says emergence IS, and where it comes from
electrical
the evidence — the dated findings, papers and results that ground it
The Chapters
how this book is read
Ch. 1
The Excess
far more parameters than the task needs
Ch. 2
Rethinking Generalization
they should overfit; they don't
Ch. 3
The Winning Ticket
tickets hidden in the slack
Ch. 4
The Implicit Bias
why SGD prefers simple
The Emergents
the theses, mechanisms, and dated findings of this theory — each a full .dlw badge (6)
Implicit bias of gradient descent2017→the optimizer as silent regularizer
ties to 0xDEADBEEFROOT0this is the cited science layer of the reserved-space theory
ENTELÉCHEIA · Book V · THE TICKET (Perísseia · περισσεία · the excess). One of the best theories of how intelligence emerges, catalogued in ROOT0's ENTELÉCHEIA universe (1991→now). Cited works belong to their authors; the cataloguing and the universe-framing are ROOT0's under the DLW standard. Kin: the anchor 0xDEADBEEF (Book I), and the sister books across ENTELÉCHEIA. Ties to ttu1 and the-seed. Domain: Artificial Intelligence.