More Like this

1. Fairness Under Composition

   Dwork, C; Ilvento, C (January 2019, 10th Innovations in Theoretical Computer Science Conference (ITCS 2019))

   Blum, A (Ed.)

   Algorithmic fairness, and in particular the fairness of scoring and classification algorithms, has become a topic of increasing social concern and has recently witnessed an explosion of research in theoretical computer science, machine learning, statistics, the social sciences, and law. Much of the literature considers the case of a single classifier (or scoring function) used once, in isolation. In this work, we initiate the study of the fairness properties of systems composed of algorithms that are fair in isolation; that is, we study fairness under composition. We identify pitfalls of naïve composition and give general constructions for fair composition, demonstrating both that classifiers that are fair in isolation do not necessarily compose into fair systems and also that seemingly unfair components may be carefully combined to construct fair systems. We focus primarily on the individual fairness setting proposed in [Dwork, Hardt, Pitassi, Reingold, Zemel, 2011], but also extend our results to a large class of group fairness definitions popular in the recent literature, exhibiting several cases in which group fairness definitions give misleading signals under composition. 

   more » « less
2. Discriminating lower mantle composition

   https://doi.org/10.1016/j.pepi.2020.106552  

   Houser, C.; Hernlund, J.W.; Valencia-Cardona, J.; Wentzcovitch, R.M. (November 2020, Physics of the Earth and Planetary Interiors)

   null (Ed.)
3. Accelerating Stochastic Composition Optimization

   Mengdi Wang, Ji Liu (January 2017, Journal of machine learning research)
4. The composition of Mars

   https://doi.org/doi:10.1016/j.gca.2020.01.011  

   Yoshizaki, Takashi; McDonough, William F. (March 2020, Geochimica et cosmochimica acta)

   Comparing compositional models of the terrestrial planets provides insights into physicochemical processes that produced planet-scale similarities and differences. The widely accepted compositional model for Mars assumes Mn and more refractory elements are in CI chondrite proportions in the planet, including Fe, Mg, and Si, which along with O make up >90% of the mass of Mars. However, recent improvements in our understandings on the composition of the solar photosphere and meteorites challenge the use of CI chondrite as an analog of Mars. Here we present an alternative model composition for Mars that avoids such an assumption and is based on data from Martian meteorites and spacecraft observations. Our modeling method was previously applied to predict the Earth’s composition. The model establishes the absolute abundances of refractory lithophile elements in the bulk silicate Mars (BSM) at 2.26 times higher than that in CI carbonaceous chondrites. Relative to this chondritic composition, Mars has a systematic depletion in moderately volatile lithophile elements as a function of their condensation temperatures. Given this finding, we constrain the abundances of siderophile and chalcophile elements in the bulkMars and its core. The Martian volatility trend is consistent with <7 wt% S in its core, which is significantly lower than that assumed in most core models (i.e., >10 wt% S). Furthermore, the occurrence of ringwoodite at the Martian core-mantle boundary might have contributed to the partitioning of O and H into the Martian core. 

   more » « less
5. The composition complexity of majority

   Lecomte; Victor; Ramakrishnan, Prasanna; Tan; Li-Yang (January 2022, Leibniz international proceedings in informatics)