Articles on transferring knowledge between sources of qualitatively known fidelity in self-driving materials laboratories, grouped in four chapters. Each chapter removes one assumption the one before relied on. Every published article links to its code and data repository and its Zenodo record. Each card also names the framework the article belongs to.
Chapter I
Sources: trusting and spending a borrowed source¶
S1 · PEGKi_Rating
Evaluating transfer across modalities in autonomous materials systems
PE·GKi PECTINATE
S2 · TENKi_Savings
TENKi: counting the value of borrowed knowledge to a materials campaign in measurements saved
TEN·Ki PECTINATE
S3 · in preparation
Pull scheduling lets a network of frugal twins share multi-fidelity campaigns
∧ PURKINJE
Chapter II
Updates: when the evidence changes¶
U1 · NablaSDL_Revision
Graded revision lets autonomous campaigns update symbolic laws of SiGeC semiconductors from mixed-fidelity evidence
∇ TRABECULAE
U2 · CircSDL_History
Processing history attached to each value separates conflicting evidence on metastable (Si)GeSn semiconductors
∘ TRABECULAE
U3 · LeadstoSDL_Reach
Provenance records predict which multi-scale conclusions a foundation-potential release can change in SiGeC semiconductors
⇝ TRABECULAE
U4 · ForallSDL_Coverage
Covering and loss-tolerant batch designs guarantee what autonomous campaigns sample in SiGeC semiconductors
∀ TRABECULAE
U5 · TurnstileSDL_Audit
Hash-committed pre-registration and exact unlearning make belief revisions in autonomous SiGeC semiconductor campaigns checkable
⊢ TRABECULAE
Chapter III
Group: when several deciders share the loop¶
G1 · QedSDL_Endpoint
How the fruit fly connectome can help autonomous materials discovery
∎ PURKINJE
G2 · HatSDL_Forecast
Laboratory plans let a self-driving database prepare for the workload of a two-dimensional materials laboratory network
x̂ PURKINJE
G3 · in preparation
Hypergraphs make team structure a controlled variable in a network of low-cost self-driving laboratories
∪ PURKINJE
Chapter IV
Owners: when the knowledge is owned by separate parties¶
O1 · MuSDL_Metering
Compound-specific memory demand determines when a research group and a computing centre benefit from sharing information in two-dimensional semiconductor screening
μ SEPTUM
O2 · EpsilonSDL_Disclosure
Minimal reciprocal disclosure in shared semiconductor screening depends on memory pressure, billing, timing masking and division of work
ε SEPTUM