J.P. Salomon Research

Research Programs

Each program crystallizes a completed body of scholarship. Explore the flagship papers, quantitative results, and frontier questions animating the next phase of work.

Biospin: Magnetosome-Based Spintronic Computing

Harnessing biological magnetosomes and magnetorheological fluids for ultra-low-power adaptive hardware.

The Biospin papers distill a hybrid architecture where magnetosome chains encode computation, magnetorheological fluids route signals, and spintronic operations achieve picojoule-level energy use. The research quantifies throughput, thermodynamic limits, and bio-integration roadmaps for edge hardware.

Flagship Completed Papers

  • Bio Inspired Robotics — magnetosome logic and MR fluids for agile, low-power actuation.
  • Emerging Robotic Movements — adaptive movement primitives powered by spin-aligned computation.

Highlighted Metrics

  • Per-operation energy: 0.01–1.0 pJ.
  • Throughput: ~109 ops/s per magnetosome cluster.
  • Gate fidelity: >95% with error correction; nanowatt-scale active power.

Frontier Questions

How quickly can magnetosome substrates scale to hybrid CPU accelerators, and what bio-engineering steps unlock stable industrial fabrication?

EVIM: Emergent Vortex-Information Model

Topological information vortices as the backbone of robust, self-organizing intelligence.

EVIM captures how information densifies into vortices that persist under sparsity, enabling systems to rebound from perturbations. The completed whitepaper defines detection algorithms, sparsity sweeps, and resilience metrics for vortex-centric AGI.

Flagship Completed Paper

  • Growth-Based AGI Whitepaper — vortex formation as a measurable path to scalable, fault-tolerant cognition.

Highlighted Metrics

  • Typical run: 3 persistent vortices with average density 0.72.
  • Robustness: vortices survive 20–30% sparsity before significant degradation.
  • Recovery reclaims 95% information flow within target thresholds.

Frontier Questions

Can vortex signatures provide control handles for online adaptation and safety in distributed AGI?

Symphonic Mind: Unified Emergent Intelligence

Neural resonance, relational topology, and emergence orchestrated into a symphonic architecture.

Symphonic Mind papers frame cognition as harmonics across neural, relational, and emergent layers. Completed studies benchmark synchronization, resonance quality, and multi-scale coherence in hybrid cognitive stacks.

Flagship Completed Papers

  • Analysis of Gemma 4B as a Language Module — large-language models as compositional instruments.
  • What Drives Intelligence & What Makes Machines Think — dialogues on emergence and neural grounding.
  • Loom: The Complete Architecture of Topological Consciousness — orchestration blueprint.

Highlighted Metrics

  • Resonance quality factors above 0.8 during learning runs.
  • Order parameters track synchrony transitions across coupling sweeps.
  • Persistent homology captures stable topological motifs during cognition.

Frontier Questions

How can harmonic tuning be automated so neural substrates, relational scaffolds, and emergent dynamics self-calibrate in real time?

Relational Intelligence Framework

Intelligence as a topology of evolving relations, decentralization, and emergent coordination.

Eight completed papers articulate a unifying theory: intelligence surfaces when relations co-adapt, topology stabilizes, and emergence is scaffolded. They provide formal models, decentralized learning experiments, and philosophical grounding.

Flagship Completed Papers

  • Intelligence as Relational Dynamics (canonical & preprint) — theoretical core.
  • The Relational Intelligence Revolution — unifies arguments across domains.
  • Emergent Intelligence as Relational Topology — multi-agent structural learning.
  • Consciousness as a Dance of Relations & Blockchain-Enabled Relational Dynamics.

Highlighted Metrics

  • Order parameters, mutual information, and network efficiency track emergence.
  • Decentralized learning benchmarks convergence time, robustness, and sample efficiency.
  • Meta-relational updates demonstrate self-evolving communication topologies.

Frontier Questions

What governance and incentives enable relational intelligence to scale without central orchestration?

Topological Equivalence (SEP)

Structural invariants under Levi lifts and parity between hypergraph-native and Levi graph learners.

The SEP program shows Levi lifts conserve key metrics, making standard GNNs competitive with hypergraph-native models. Completed analyses benchmark synthetic and real datasets with statistical confidence.

Flagship Completed Paper

  • 1706.03762 (Annotated) — structural equivalence foundations.

Highlighted Metrics

  • Spectral distance after Levi lift: ≈0.12; Wasserstein incidence distance: ≈0.05.
  • Downstream parity: HGNN/HyperGCN converge with GCN/GIN across DBLP and synthetic sweeps.
  • Confidence: 5+ seeds with 95% CIs on performance metrics.

Frontier Questions

How does SEP extend to higher-order dynamical tasks and continuous-time hypergraph processes?