This model maps weights to ternary values {-1, 0, 1} (aka trits). One trit holds log(3)/log(2) ≈ 1.58 bits of information. To represent a single trit by itself would require 2 bits, but it is possible to pack 5 trits into 8 bits. This article explains it well: https://compilade.net/blog/ternary-packing
By using 4 ternary weights per 8 bits, the model is not quite as space-efficient as it could be in terms of information density. (4*1.58)/8 = 0.79 vs (5*1.58)/8 = 0.988 There is currently no hardware acceleration for doing operations on 5 trits packed into 8 bits, so the weights have to be packed and unpacked in software. Packing 5 weights into 8 bits requires slower, more complex packing/unpacking algorithms.
My neighborhood has a vehicle gate that is opened by an RF clicker, and a pedestrian gate opened by an RFID tag. I copied both of mine to my flipper. A couple months ago the coin cell battery in my clicker died, but I had my flipper! I also use the cloned RFID tag fairly regularly. There are two RFID tags in the house and more than two people using them, so I use my flipper for that too.
Transposons and pseudogenes may be nonfunctional for an individual organism, but aren't they useful for evolution to create new genes? Wikipedia suggests some usefulness:
"Pseudogene sequences may be transcribed into RNA at low levels, due to promoter elements inherited from the ancestral gene or arising by new mutations. Although most of these transcripts will have no more functional significance than chance transcripts from other parts of the genome, some have given rise to beneficial regulatory RNAs and new proteins." (https://en.wikipedia.org/wiki/Pseudogene)
I would define truly junk DNA as any DNA that if removed would be beneficial or not harmful to the majority of organisms in a species, and be beneficial or not harmful to the ability of a species to adapt and evolve its genome.
By using 4 ternary weights per 8 bits, the model is not quite as space-efficient as it could be in terms of information density. (4*1.58)/8 = 0.79 vs (5*1.58)/8 = 0.988 There is currently no hardware acceleration for doing operations on 5 trits packed into 8 bits, so the weights have to be packed and unpacked in software. Packing 5 weights into 8 bits requires slower, more complex packing/unpacking algorithms.