Instituto Geológico

New genera and species of lungfish (Sarcopterygii, Dipnoi) from the Upper Jurassic of western Gondwana, in the Jatobá basin, northeastern Brazil

The present study describes three new species and two new genera of lungfish from the Upper Jurassic (Tithonian) of the Aliança Formation (Jatobá Basin), in the state of Pernambuco, northeastern Brazil. The new taxa, Ceratodus ibimiriensis sp. nov., Tekoaceratodus pernambucensis...

Geochemical fingerprints and tectonic setting of the Late Palaeozoic marginal basin of the Bariloche Complex (SW Gondwana), North Patagonian Andes of Argentina

The understanding of the Late Palaeozoic evolution of the south-western Gondwana margin is limited due to a lack of outcrops and data. New insights were achieved by fieldwork, petrographic characterization and geochemical fingerprinting on metasedimentary and interlayered metabas...

A Two-Space Scale-Invariant Direct Multisearch Method for Mixed-Variable Multiobjective Optimization

Many multiobjective black-box problems combine continuous, ordered-discrete, and categorical variables. Direct search on such domains requires a geometry that respects each variable type and is insensitive to arbitrary scaling, spacing, and labeling. We propose DMS-SI-Mix, a dete...

Mixed-Variable Optimisation as a Metric Product Space: Transient Categorical Geometry and a Hierarchy of Local Optimality

v2, August 2026: substantially revised version — restructured into eight sections, renamed optimality hierarchy (single-ordering / visited-orderings / all-orderings), corrected algorithmic section, and a new controlled computational experiment. A version of this manuscript is und...

Mixed-Variable Optimisation as a Metric Product Space: Transient Categorical Geometry and a Hierarchy of Local Optimality

v2, August 2026: substantially revised version — restructured into eight sections, renamed optimality hierarchy (single-ordering / visited-orderings / all-orderings), corrected algorithmic section, and a new controlled computational experiment. A version of this manuscript is und...

A Two-Space Scale-Invariant Direct Multisearch Method for Mixed-Variable Multiobjective Optimization

Many multiobjective black-box problems combine continuous, ordered-discrete, and categorical variables. Direct search on such domains requires a geometry that respects each variable type and is insensitive to arbitrary scaling, spacing, and labeling. We propose DMS-SI-Mix, a dete...