Does rolling back a robust topology-optimisation step that spikes compliance, and redoing it at half the move, beat shrinking the move afterwards?

What it found

Follow-up to simdesign-robustadapt. The same nine robust half-MBB cases are run with five move-limit rules (45 deterministic runs): fixed 0.2, 0.1, 0.05, the earlier adaptive rule, and reject-and-retry, which discards a step whose worst-case compliance jumps more than 3× and redoes it at half the move. Blueprint failures: fixed 0.2, 5 of 9; adaptive, 2; retry, 1; fixed 0.1 and 0.05, none. Retry costs at most 2% more evaluations but never beat fixed 0.05, and on eroded-design quality it is 1.22× the best on average (fixed 0.1 hides one eroded design 25× worse that the blueprint test passes). Stylised: 2D dimensionless, one problem, one run per cell.

A figure from its own output

Paper, code and results

• README and method
• Experiments and results
• Code
• Tests

More in Simulation-first design

• Can penalty continuation free SIMP topology-optimisation starts that stay stuck in a different local optimum?Can penalty continuation free SIMP starts stuck in a different local optimum? No: it sends all 5 stuck starts of 20 to one design (c 204.01) worse than the 202.83 three reach by running longer at p = 3. Stylised.
• Is the start-dependence of fixed-penalty SIMP topology optimisation just slow convergence?Is the start-dependence of fixed-penalty SIMP just slow convergence? Over 1200 iterations the median of 20 random starts reaches the uniform-start design (0.0% of elements differ, from 3.9% at 150), but the compliance spread only falls 1.43% → 0.85% (stylised, 2D, one mesh).
• How much does a SIMP topology optimisation depend on its starting density field, and does penalty continuation remove it?How much does a SIMP topology optimisation depend on its starting density field? With a fixed penalty, final compliance spreads 1.4–2.8% over 30 random starts and 2–23% of elements change state; with penalty continuation the spread is 0–0.4% (stylised, 2D, 150 iterations).
• How short can SIMP penalty continuation be and still remove the dependence on the starting density field?How short can SIMP penalty continuation be and still remove the start dependence? A ramp of 25 iterations or fewer leaves much of it (worst start differs in 16–23% of elements, 27–31% for a fixed penalty); 50 iterations cuts compliance spread to 0.01–0.23%, and a 112-iteration ramp is most repeatable but ends at a worse compliance (stylised, 2D, 150 iterations).
• When does a cheaper beam model pick the same design as 3D finite elements?Cantilever sizing with Euler–Bernoulli, Timoshenko, 2D and 3D finite elements: stress agrees everywhere, deflection differs; cheap models land within ~5% of the 3D depth, unsafe when narrow and stubby (Euler–Bernoulli) and oversized when wide (plane stress). Stylised, assumed material.
• How do design proposals from a language-model agent compare with CMA-ES, Bayesian optimisation and random search at an equal simulation budget?Session-agent design proposals vs CMA-ES, Bayesian optimisation and random search on a bracket sized by finite elements, 40 evaluations each: agent 23.0 g vs 32–39 g baseline medians (best known 22.4 g); the agent had prior knowledge, n = 1, not an LLM benchmark. Stylised.

All 64 Simulation-first design projects →

Other research topics

Proper scoring & elicitation28Effort & peer prediction9Verification games & slashing19Detection & tolerances14Markets as routers13Prediction-market design9Paying contributors8Decentralised training systems23Digital twins & sim-to-real164LLM chip design loops40Robot manufacturing loops53Ambient AI assistant46Self-replicating manufacturing34Simulation-first design64Engineered clothing45Task-specific AI accelerators40Algorithmic chip-design loops42Open-source purpose-built robots44The software-engineering loop43Adversarial human imitation41Open-source residential construction41Autonomous multi-angle capture34Sensing, imaging and 3D31Energy supply and dense energy storage39Robot societies among people31Agentic co-design of low-cost robots27Laundry-hanging robots25Machine learning on qualitative data23Robots on the building site27Civilizations & group psychology1History & human sciences1

Produced with AI assistance (Claude). Results are from the code in this repository and are simulated or stylised unless the abstract says otherwise; nothing here implies affiliation with or endorsement by any lab, researcher or company cited. MIT licensed.