Abstract: A computing platform may be configured to (i) train an initial model object for a data science model using a machine learning process, (ii) determine that the initial model object exhibits a threshold level of bias, and (iii) thereafter produce an updated version of the initial model object having mitigated bias by (a) identifying a subset of the initial model object's set of input variables that are to be replaced by transformations, (b) producing a post-processed model object by replacing each respective input variable in the identified subset with a respective transformation of the respective input variable that has one or more unknown parameters, (c) producing a parameterized family of the post-processed model object, and (d) selecting, from the parameterized family of the post-processed model object, one given version of the post-processed model object to use as the updated version of the initial model object for the data science model.

Abstract: Provided is a system providing a solution for improving prescription drug compliance monitoring and compliance. Prescription drug compliance monitoring may be achieved by using patient's electronic device to initiate secure digital imaging. Medications may be dispensed in containers with pre-printed encoded unique identifiers and markings. The patient may use the digital camera in their smart phone to take a picture of the dispensed blister pack which may be processed and analyzed. Analysis may include assessment of the number of pills taken from and remaining in the uniquely identified blister pack. Package identification and consumption information may then be linked with other prescription information to determine if the observed rate of use is compliant with the legal prescription and to document that the imaged medication package is the same as the dispensed medication package. Data may also be linked with pharmacologic and clinical data to categorize risk and urgency of misuse.

Abstract: A framework for interpreting machine learning models is proposed that utilizes interpretability methods to determine the contribution of groups of input variables to the output of the model. Input variables are grouped based on dependencies with other input variables. The groups are identified by processing a training data set with a clustering algorithm. Once the groups of input variables are defined, scores related to each group of input variables for a given instance of the input vector processed by the model are calculated according to one or more algorithms. The algorithms can utilize group Partial Dependence Plot (PDP) values, Shapley Additive Explanations (SHAP) values, and Banzhaf values, and their extensions among others, and a score for each group can be calculated for a given instance of an input vector per group. These scores can then be sorted, ranked, and then combined into one hybrid ranking.

Abstract: A method, system, and computer-readable medium are disclosed for detecting and mitigating bias in a trained machine learning model. The method includes the steps of: training the model based on a training data set; detecting bias in the model relative to a protected class; identifying one or more groups of input variables that contribute to the bias; and mitigating bias in the model. Mitigating the bias is performed by constructing a post-processed score function that either (a) neutralizes or partially neutralizes one or more groups of input variables in the input vector of the model, or (b) utilizes a fair score approximation of the model to project the distributions for the protected class and/or the unprotected class to substantially match. In an embodiment, detecting bias in the trained model is performed by comparing distribution for two or more subpopulations based on a distance metric, such as a Wasserstein distance.

Abstract: A framework for interpreting machine learning models is proposed that utilizes interpretability methods to determine the contribution of groups of input variables to the output of the model. Input variables are grouped based on correlation with other input variables. The groups are identified by processing a training data set with a clustering algorithm. Once the groups of input variables are defined, partial dependent plot (PDP) tables for each group are calculated and stored in a memory, which are used for calculating scores related to each group of input variables for a given instance of the input vector processed by the model. Furthermore, Shapley Additive Explanations (SHAP) values for each group can be calculated by summing the SHAP values of the input variables for a given instance of an input vector per group. These scores can then be sorted, ranked for each interpretability method, and then combined into one hybrid ranking.

Abstract: A high pressure hydraulic fitting (300) includes a body (302) having a central passage (322), a channel (304), a threaded section (314), and an extension section (310), wherein the threaded section (314) is disposed between the channel (304) and the extension section (314), and wherein the central passage (322) fluidly connects a first distal end (303) with a second distal end (316) of the body (302). A first o-ring (306) is in the channel (304). A locking nut (324) is engaged to the body (302). The locking nut (324) has an internal threaded section (326) that is arranged to mate with the threaded section (314) on the body (302).

Abstract: A fuel system for a diesel engine may have a fuel tank, a fuel pump, a fuel filter module, and a fuel rail. The fuel filter module has a vacuum side and a pressure side. The vacuum side is connected to the fuel tank and to the inlet side of the fuel pump. The pressure side is connected to the outlet side of the fuel pump and to the fuel rail. The fuel filter module has a pressure regulator and a filter element in the pressure side. The pressure regulator is disposed between the outlet side of the fuel pump and the filter element. The pressure regulator recirculates unfiltered fuel to the vacuum side of the fuel filter module.

Abstract: A method and apparatus for adjusting fuel injection timing detects (401) a change in the number of fuel injections for a combustion chamber (105). During a timing adjustment period, the timing for fuel injection(s) may be adjusted (405) incrementally, e.g., in a series of steps or increments, and/or based at least in part on the time elapsed since the detected of the change in the number of fuel injections.

Abstract: A fluid pressure near a filter (103) is measured and compared (307) to a predetermined value that is based on one or more engine operating parameters. When the measured fluid pressure deviates enough from the predetermined value by an established amount, an indication of the deviation may be provided to notify of a potential problem with the filter (103), such as obstruction, restriction, or some form of clogging.