Renormalized Cellular Intelligence: When Does Model Reduction Preserve Collective Capability?
Organizations: Pfizer Inc. · Fudan University · Cornell University · Independent researcher
Abstract
As multi-agent systems grow in size and complexity, understanding which features of their interactions support collective capability becomes increasingly difficult. Can a simpler description preserve both task performance and the measured contribution of interactions between agents? Preserving collective dynamics alone is insufficient: reduction can alter decision-relevant information or give a non-communicating baseline access to pooled observations, distorting the measured advantage of interaction. We introduce Renormalized Cellular Intelligence (RCI), a framework for specifying and auditing capability preservation across scales. RCI treats the task, interacting dynamics, matched baseline, decision interfaces, and resource budget as a joint object of reduction. It specifies dynamical closure, decision compatibility, and preservation of the intended baseline comparison, and bounds the resulting error in relative capability gain. An exact analytical construction demonstrates task-score preservation, while a counterexample shows how closed reduced dynamics can erase the measured interaction gain. Using cellular systems as models of locally interacting agents, we evaluate three reduction candidates on five frozen learned models without retraining. The final candidate reduces the evolving state from eight to five dimensions per cell (37.5%). Within-run uncertainty bounds support relative gain above a prespecified threshold for the fixed model collection, yet the candidate fails the closure and decision-preservation requirements for a complete RCI certificate. RCI provides explicit criteria for distinguishing positive interaction gains from
Figures & tables
| System | State scalars | State factor | Raw ports | Site MLPs/step |
|---|---|---|---|---|
| Microscopic | 8,192 | 1,024 | 1,024 | |
| v1 | 2,048 | 1,024 | 1,024 | |
| v2 | 4,096 | 1,024 | 1,024 | |
| v3 | 5,120 | 1,024 | 1,024 |
| Candidate | Overall | |||||
|---|---|---|---|---|---|---|
| Block mean | 0.1362 | 0.1867 | 0.1371 | 0.0659 | 0.1198 | Fail |
| Four private dims | 0.1180 | 0.1153 | 0.3064 | -0.1059 | 0.2406 | Fail |
| Five private dims | 0.0720 | 0.0753 | 0.1955 | 0.0069 | 0.1666 | Fail |
| Seed | Full cert. | ||||||
|---|---|---|---|---|---|---|---|
| 1000 | 0.0469 | 0.0558 | 0.0284 | 0.0195 | 0.2421 | 0.0944 | Fail |
| 1001 | 0.0779 | 0.0818 | 0.1622 | 0.1527 | 0.1278 | -0.2309 | Fail |
| 1002 | 0.0540 | 0.0559 | 0.1074 | 0.0667 | 0.2219 | -0.0346 | Fail |
| 1003 | 0.0460 | 0.0494 | 0.0366 | 0.0028 | 0.3756 | 0.2375 | Fail |
| 1004 | 0.0725 | 0.0706 | 0.0949 | 0.1727 | 0.1220 | -0.1466 | Fail |
Appendix figures & tables16 assets
Supplementary material from the paper’s appendix.
Appendix
| Control | Preserved / changed | Evidence and interpretation |
|---|---|---|
| Self-copy null | Same samples, states, read noise, decoder, deadline; | Observed in all inference probes; communication gain only. |
| Episode-misaligned channel | Endpoint channel marginal and spatial dependence; wrong task instance | Observed in Probe 1; checks task alignment, not a per-round fault process. |
| Role-placement pair | Graph, size, role counts, dynamics and noise; information locations | Observed; isolates location under fixed response mechanism. |
| Direct centralized agent | Full observation access and declared compute budget | Not measured; no superiority claim over centralized inference. |
| Serial same decomposition | Same subproblems, information, total work, and decoder | Not measured; no parallel-computation advantage claim. |
| Message/energy matched null | Equal traffic, precision, energy, and useful compute | Not measured; equal update weights are insufficient. |
| Study | Stated object or demonstration | Question posed by the present RCI contract |
|---|---|---|
| Dacre et al. (2025) | Local sensing and global geometric estimation in a physical sensor array | Does a reduced state preserve the declared task contrast under the same observation and null interfaces? |
| Schmidt et al. (2024) | Graph structure, spectra, and diffusion under coarse-graining/rescaling | Does preserved effective dynamics also retain decisions and the matched-null task score? |
| Etcheverry et al. (2026) | Cellular reasoning with local computation and asynchronous updates | Which state reductions preserve capability at a fixed task deadline and budget? |
| Framework | Established object | RCI emphasis |
|---|---|---|
| Information bottleneck / information-theoretic RG | Relevant information and coarse degrees of freedom | Task decisions and an interaction–null contrast. |
| State abstraction / homomorphisms | Related dynamics, rewards, and value functions | Two experimental arms and their admissible observation interfaces. |
| Causal abstraction | Consistent predictions under mapped interventions | Communication removal with preserved decoder, weighting, and budget meaning. |
| Distributed inference | Propagation and collective estimation | Capability relative to local information access and a practical threshold. |
| Case | Supported mechanism or diagnostic | Relation to RCI |
|---|---|---|
| Inference 1 | Information placement and task alignment matter beyond general propagation summaries | Identifies distinctions a useful description must retain. |
| Inference 2 | Co-reachability predicts within-family gain under frozen evaluation | Supports task-conditioned response compression. |
| Inference 3 | Finite calibration works; response-moment transfer fails across distributions | Separates measurability from descriptor sufficiency. |
| Adaptive cellular pilot | Inheritance changes harvest and persistent behavioral novelty | Motivates adaptive effective units; does not measure new-task capability. |
| Condition | Capability gain | Agreement |
|---|---|---|
| Random roles | 0.6590 | |
| Segregated roles | 0.5011 | |
| Random roles, misaligned channel | 0.6596 |
| Held-out change | Old-descriptor RMSE | Co-reachability RMSE | |
|---|---|---|---|
| Layout | 0.05758 | 0.00355 | 0.9975 |
| Horizon | 0.06378 | 0.00345 | 0.9984 |
| Noise | 0.09599 | 0.00557 | 0.9967 |
| Topology and size | 0.07378 | 0.00283 | 0.9989 |
| Joint | 0.10843 | 0.00951 | 0.9932 |
| Calibration | Gaussian | latent + Gaussian noise | Binary observation |
|---|---|---|---|
| Ideal | 0.00358 | 0.13114 | 0.12215 |
| 32 rounds | 0.00575 | 0.13164 | 0.12011 |
| 128 rounds | 0.00359 | 0.13104 | 0.12252 |
| 512 rounds | 0.00345 | 0.13129 | 0.12118 |
| Quantity | Mean [interval] | |
|---|---|---|
| Maximum early harvest | ||
| Maximum late novelty rate | ||
| Learned minus random early harvest | ||
| Learned minus random late novelty |
| Claim | Main artifact | Status |
|---|---|---|
| Topology-summary collision | rci_probe/results/summary.json ; stress.csv | Observed and analytically diagnosed |
| Frozen co-reachability transfer | rci_probe2/results/metrics.csv ; frozen.json | Observed within Gaussian family |
| Finite calibration / failed transfer | rci_probe3/results/metrics.csv ; matched_moment_pairs.csv | Observed, no refit |
| Inheritance and novelty | paper1_pilot/condition_summary.csv ; paired_contrasts.csv | Exploratory finite-time proxies |
| Analytical reduction | Section 3.3.1 and its derivation | Exact symbolic result |
| Frozen cellular state reductions | reductions/ (one directory per candidate), primary_result.json | Evaluated; all candidates retained |
| Seed | Selected update | Historical validation error |
|---|---|---|
| 1000 | 23,900 | 0.116847 |
| 1001 | 22,900 | 0.075730 |
| 1002 | 24,000 | 0.109782 |
| 1003 | 23,600 | 0.111996 |
| 1004 | 22,900 | 0.080101 |
| 8 | 0.055110 | 0.059756 | 0.017473 | 0.016744 | 0.020920 | 0.014443 |
|---|---|---|---|---|---|---|
| 16 | 0.057891 | 0.061872 | 0.036579 | 0.061577 | 0.076409 | 0.059628 |
| 24 | 0.059395 | 0.062639 | 0.068769 | 0.090185 | 0.184230 | 0.142408 |
| 32 | 0.059428 | 0.062696 | 0.085890 | 0.082877 | 0.257976 | 0.217890 |
Multi-Agent Systems
Multi-Agent Systems studies how autonomous agents coordinate, compete, communicate, and learn in shared or interdependent tasks.
See all 709 papers