Abstract: One embodiment of the invention provides a method for federated learning (FL) comprising training a machine learning (ML) model collaboratively by initiating a round of FL across data parties. Each data party is allocated tokens to utilize during the training. The method further comprises maintaining, for each data party, a corresponding data usage profile indicative of an amount of data the data party consumed during the training and a corresponding participation profile indicative of an amount of data the data party provided during the training. The method further comprises selectively allocating new tokens to the data parties based on each participation profile maintained, selectively allocating additional new tokens to the data parties based on each data usage profile maintained, and reimbursing one or more tokens utilized during the training to the data parties based on one or more measurements of accuracy of the ML model.

Abstract: Systems and methods of the present disclosure enable efficient dataset sampling by using one or more processors for receiving a sample request including a dataset, a target feature, and a sample size to be produced. The processor(s) determines, from a library of smart sampling algorithms based on the sample request, a smart sampling algorithm associated with a sampling parameter that satisfies the sample request. The processor(s) configures the sampling parameter of the smart sampling algorithm of a smart sampling engine to obtain a configured smart sampling engine to generate a smart sampled dataset that represents the target feature and satisfies the sample size. The processor(s) inputs the dataset into the configured smart sampling engine so as to cause the configured smart sampling engine to execute the smart sampling algorithm and output the smart sampled dataset, and returns the smart sampled dataset to the computing device.

Abstract: Disclosed are systems and methods for scenario planning by using specially programmed software engines to simulate and detect particular feature variations leading to particular outcomes based on modeling with machine learning techniques. The disclosed technology enable improved model debugging, improved simulation efficiency and accuracy, improved model explainability, improved identification of high risk or high reward scenarios, among other improvements and combinations thereof. In some embodiments, the disclosed technology implements computerized optimization techniques applied via variation generation across a dataset of test input records to optimize for feature variation along with outcome variation.

Abstract: Systems and methods for scenario planning include using specially programmed software engines to simulate and detect particular feature variations leading to particular outcomes based on modeling with machine learning techniques. The systems and methods improve model debugging, simulation efficiency and accuracy, model explainability, identification of high risk or high reward scenarios, among other improvements and combinations thereof. The systems and methods implement computerized optimization techniques applied via variation generation across a dataset of test input records to optimize for feature variation along with outcome variation. Moreover, the systems and methods may provide and/or realize a minimized variation to input data that correspond to a point of transition from one state to another state in an outcome that results from the input data, where the transition to another state is termed a “significant” variation to the output data.

Abstract: One or more computer processors determine a tolerance value, and a norm value associated with an untrusted model and an adversarial training method. The one or more computer processors generate a plurality of interpolated adversarial images ranging between a pair of images utilizing the adversarial training method, wherein each image in the pair of images is from a different class. The one or more computer processors detect a backdoor associated with the untrusted model utilizing the generated plurality of interpolated adversarial images. The one or more computer processors harden the untrusted model by training the untrusted model with the generated plurality of interpolated adversarial images.

Abstract: Disclosed are systems and methods for scenario planning by using specially programmed software engines to simulate and detect particular feature variations leading to particular outcomes based on modeling with machine learning techniques. The disclosed technology enable improved model debugging, improved simulation efficiency and accuracy, improved model explainability, improved identification of high risk or high reward scenarios, among other improvements and combinations thereof. In some embodiments, the disclosed technology implements computerized optimization techniques applied via variation generation across a dataset of test input records to optimize for feature variation along with outcome variation.

Abstract: Systems and methods are disclosed for scenario planning by using specially programmed software engines to simulate and detect particular feature variations leading to particular outcomes based on modeling with machine learning techniques. The disclosed technology enables improved model debugging, improved simulation efficiency and accuracy, improved model explainability, improved identification of high risk or high reward scenarios, among other improvements and combinations thereof. Computerized optimization techniques applied via variation generation across a dataset of test input records enable optimization for feature variation along with outcome variation. Moreover, the disclosed techniques may provide and/or realize a minimized variation to input data that correspond to a point of transition from one state to another state in an outcome that results from the input data, where the transition to another state is termed a “significant” variation to the output data.

Abstract: Disclosed are systems and methods for scenario planning by using specially programmed software engines to simulate and detect particular feature variations leading to particular outcomes based on modeling with machine learning techniques. The disclosed technology enables improved model debugging, improved simulation efficiency and accuracy, improved model explainability, improved identification of high risk or high reward scenarios, among other improvements and combinations thereof. In some embodiments, the disclosed technology implements computerized optimization techniques applied via variation generation across a dataset of test input records to optimize for feature variation along with outcome variation.

Abstract: Systems and methods for smart instance selection in accordance with embodiments of the invention are illustrated. One embodiment includes a system for selecting explanatory instances in datasets, including a processor, and a memory, the memory containing an instance selection application that configures the processor to: obtain a dataset comprising a plurality of records, obtain a machine learning model configured to classify records, initialize an explainer model, select at least one key instance from the dataset estimated to have explanatory power when provided to the explainer model, provide the explainer model with the selected at least one key instance; and provide an explanation produced by the explainer model.

Abstract: One embodiment of the invention provides a method for federated learning (FL) comprising training a machine learning (ML) model collaboratively by initiating a round of FL across data parties. Each data party is allocated tokens to utilize during the training. The method further comprises maintaining, for each data party, a corresponding data usage profile indicative of an amount of data the data party consumed during the training and a corresponding participation profile indicative of an amount of data the data party provided during the training. The method further comprises selectively allocating new tokens to the data parties based on each participation profile maintained, selectively allocating additional new tokens to the data parties based on each data usage profile maintained, and reimbursing one or more tokens utilized during the training to the data parties based on one or more measurements of accuracy of the ML model.

Abstract: An approach for providing prediction and optimization of an adversarial machine-learning model is disclosed. The approach can comprise of a training method for a defender that determines the optimal amount of adversarial training that would prevent the task optimization model from taking wrong decisions caused by an adversarial attack from the input into the model within the simultaneous predict and optimization framework. Essentially, the approach would train a robust model via adversarial training. Based on the robust training model, the user can mitigate against potential threats by (adversarial noise in the task-based optimization model) based on the given inputs from the machine learning prediction that was produced by an input.

Abstract: One or more computer processors determine a tolerance value, and a norm value associated with an untrusted model and an adversarial training method. The one or more computer processors generate a plurality of interpolated adversarial images ranging between a pair of images utilizing the adversarial training method, wherein each image in the pair of images is from a different class. The one or more computer processors detect a backdoor associated with the untrusted model utilizing the generated plurality of interpolated adversarial images. The one or more computer processors harden the untrusted model by training the untrusted model with the generated plurality of interpolated adversarial images.