Abstract: The present disclosure describes techniques for determining performance of a classifier. A first machine learning model and a second machine learning model may be trained by aggregating updates to the first machine learning model and the second machine learning model received from a plurality of client computing devices. A cumulative distribution function (CDF) associated with a distribution of the positive samples in the user data may be estimated using the trained first machine learning model. A probability density function (PDF) associated with a distribution of the negative samples in the user data may be estimated using the trained second machine learning model. An integration-based computation of an area under the receiver operating characteristic curve (AUC) of the classifier may be performed using the PDF and the CDF.

Abstract: Provided are new classes of reconfigurable, multistable metasurfaces using the physics of colloids confined within nematic liquid crystals (NLCs). The colloids are physically moved from one stable location to another as defined by NLC-colloid interactions by external switching fields. These colloids can be scatterers within reconfigurable metasurfaces with electromagnetic (EM) responses that change with colloid location. because the colloids are moved between stable loci, these metasurfaces are also multistable, and require energy input only when changing colloid position. Furthermore, because the colloids can be returned to their original positions by simply reversing the switching field, all changes in EM responses are entirely reversible.

Abstract: Provided are compositions that include a nematic colloid, the nematic colloid comprising a nematic liquid crystal and a key colloid; and a lock colloid, the lock colloid optionally having at least two arms extending therefrom, the lock colloid being configured for assembly with the key colloid of the nematic colloid, the assembly optionally being mediated by a dipole interaction between the colloid and the lock colloid, by a disinclination line of the nematic colloid, or any combination thereof. Also provided are related methods. The disclosed compositions and methods can be used to, e.g., assemble chain and lattice structures from the key colloids by exploiting disinclination lines and dipole defects of the components of the compositions.

Abstract: The present invention relates to a formulation comprising a PD-L1/LAG-3 bispecific antibody, and particularly to a pharmaceutical formulation comprising a PD-L1/LAG-3 bispecific antibody, a buffer, a stabilizer, a surfactant and a chelating agent. Furthermore, the present invention further relates to therapeutic or prophylactic use of these formulations.

Abstract: Provided are compositions that include a nematic colloid, the nematic colloid comprising a nematic liquid crystal and a key colloid; and a lock colloid, the lock colloid optionally having at least two arms extending therefrom, the lock colloid being configured for assembly with the key colloid of the nematic colloid, the assembly optionally being mediated by a dipole interaction between the colloid and the lock colloid, by a disinclination line of the nematic colloid, or any combination thereof. Also provided are related methods. The disclosed compositions and methods can be used to, e.g., assemble chain and lattice structures from the key colloids by exploiting disinclination lines and dipole defects of the components of the compositions.

Abstract: The present invention relates to formulations comprising an anti-LAG-3 antibody, and in particular to pharmaceutical formulations comprising an antibody specifically binding to LAG-3 molecules, a buffer, a stabilizer and a surfactant. Furthermore, the present invention also relates to therapeutic or prophylactic use of these formulations.