Karan Diwan

Gate-Level Cybernetic Classifier

A binary classifier that corrects its own decision boundary from feedback, in pure deterministic hardware.

This project asks how far adaptive behavior can go without software: can a purely deterministic gate-level circuit alter its own decision boundary in response to feedback, with no processor or instruction set involved? It evolves in three stages, from strict Boolean pattern matching, through pop-count similarity against a variable threshold, to a feedback-driven system that corrects itself.

Cybernetic classifier MIDS
Cybernetic self-learning adaptive system implementing MIDS algorithm - autonomously altering its decision boundary while detecting and supressing the timing hazard.

Versions built

Cybernetic classifier block diagram, v1.0
Detector v1.0: a manually alterable decision boundary, the precursor to autonomous correction.

Two correction algorithms

Version 1.1 implements two distinct approaches to threshold correction, each with a formal convergence proof for thresholds M in the range 1 to 15: Max-Initialized Decremental Search (MIDS), which starts at the maximum threshold and searches downward, and State-Aware Threshold Update (SATU), which computes the correct threshold directly from the current and desired output.

PropertyMIDSSATU
Correction speedO(N)O(1)
StrategyIterative threshold traversalDirect threshold computation
Threshold memoryRequiredNot required
SynchronizationMultiple requirementsSingle synchronizer
Hardware complexityHigherLower
Correction speed complexity comparison chart
Correction speed: O(N) search versus O(1) direct computation.
Toolchain: Logisim, Icarus Verilog, Verilator, Yosys, OpenSTA, GTKWave, Sky130HD
View repository on GitHub →
← Hopfield Network Next: FIFO Architectures →