Explore data, tools and research questions. Leave with a plan you can use with your AI.
Browse, compare and export. No account needed.
Choose a question. Adapt the plan. Take it to your AI.
How misleading can a nominal 95% success-rate interval be when only twenty independent tasks are observed?
Do evaluation harnesses agree on the same frozen predictions once normalization, aggregation and failure handling are made explicit?
Can another operator reconstruct one bounded agent-evaluation attempt from its task, environment, actions, failures and scoring rule?
Upstream project repository
Multidimensional array operations and numerical linear algebra for scientific Python.
Investigate How do sequential and pairwise array reductions behave on cancellation-heavy inputs?
UCI Machine Learning Repository
Flower measurements for a compact, interpretable multiclass baseline and leakage audit.
Investigate How much do iris classification results change when preprocessing is restricted to training folds?
Executable Science local pilot drafts
Enumerate all binomial outcomes at n=20 and p=0.05, then compare exact interval-coverage calculations with a fixed-seed simulation.
Investigate Do nominal 95% intervals achieve that coverage in this prespecified small-sample setting?
Upstream project repository
Property-based testing tools that generate and simplify test inputs.
Investigate Can a fixed property-based test expose a specified floating-point overclaim within a bounded search protocol?
scikit-learn documentation
A guide to model-evaluation splits, including grouped and time-series data, nested model selection and permutation-based assessment.
Investigate How much optimism arises when repeated observations from the same group are split across train and test?
Upstream project repository
Numerical routines for integration, differential equations, optimization, transforms, and statistical analysis.
Investigate Which declared tolerances are attained by adaptive quadrature on a fixed family of endpoint-singular integrals?
Curated starting points. Library inclusion does not mean execution, independent verification or publication here.
Choose inputs, comparisons and metrics before trusting a score.
Explore the collection →02Understand small samples and the assumptions behind an apparent improvement.
Explore the collection →03Trace published work to its sources and a useful next check.
Explore the collection →Use a brief with your preferred AI tool, develop your own work, or explore our journal’s approach to inspectable research.
We’re building a lasting home for inspectable research. Start a private submission while the journal prepares for public launch.