Introducing CMG, a coordinate-wise gating mechanism for quantum-inspired sequence learning. Achieves 91%+ improvement over scalar gating in multi-step quantum dynamics forecasting, maintaining prediction error at 10^-4 scale across 4-16 step horizons.
#QuantumML #SequenceLearning #Research
#QuantumML #SequenceLearning #Research
Complementary Matrix Gating for Quantum-Inspired Fast-Weight Programmers in Quantum Dynamics Forecasting
iq.fp2.dev
July 31, 2026 at 2:54 AM
Introducing CMG, a coordinate-wise gating mechanism for quantum-inspired sequence learning. Achieves 91%+ improvement over scalar gating in multi-step quantum dynamics forecasting, maintaining prediction error at 10^-4 scale across 4-16 step horizons.
#QuantumML #SequenceLearning #Research
#QuantumML #SequenceLearning #Research
Spiking SDM sequence machine (2007) and transformer (2017) instantiate the same five operations: my preprint documenting the correspondence between the two (the first is based on my PhD thesis) is now on arxiv.
arxiv.org/abs/2605.00662
#transformers #spikingneurons #sequencelearning #rankorder
arxiv.org/abs/2605.00662
#transformers #spikingneurons #sequencelearning #rankorder
Spiking Sequence Machines and Transformers
Sequence learning reduces to similarity-based retrieval over a temporally indexed representation space, a constraint on any sequence model, not a property of a specific architecture. We show that a sp...
arxiv.org
May 5, 2026 at 2:35 AM
Spiking SDM sequence machine (2007) and transformer (2017) instantiate the same five operations: my preprint documenting the correspondence between the two (the first is based on my PhD thesis) is now on arxiv.
arxiv.org/abs/2605.00662
#transformers #spikingneurons #sequencelearning #rankorder
arxiv.org/abs/2605.00662
#transformers #spikingneurons #sequencelearning #rankorder