Abstract: Methods, systems, and computer program products for automatically modifying responses from generative models using artificial intelligence techniques are provided herein. A computer-implemented method includes obtaining data pertaining to at least one conversation involving at least one automated conversation exchange software program and at least one user; identifying, among words proposed by the at least one automated conversation exchange software program in connection with the at least one conversation, words qualifying as belonging to one or more predetermined categories by processing the obtained data using artificial intelligence techniques; determining, by processing the identified words and at least one word-based data source, one or more alternate words; modifying at least a portion of the proposed words by replacing at least a portion of the identified words with at least a portion of the one or more alternate words; and performing at least one automated action based on the modifying.

Abstract: Methods, systems, and computer program products for active explanation guided learning are provided herein. A computer-implemented method includes identifying a subset of training examples, from a set of training examples, based on at least one of (i) an uncertainty metric computed for each one of the training examples and (ii) an influence metric computed for each one of the training examples; outputting said subset of training examples to a user; obtaining, from the user, a user explanation for each training example in said subset of training examples, wherein each of the user explanations identifies at least one part of the corresponding training example; and training a machine learning model based at least in part on the user explanations, wherein said training comprises prioritizing the identified parts of the training examples in the subset.

Abstract: One embodiment provides a method, including: receiving a dataset and a model corresponding to a bias checker, wherein the bias checker detects bias within both the dataset and the model, based upon a bias checking algorithm and a bias checking policy, wherein the dataset comprises a plurality of attributes; testing the bias checking algorithm of the bias checker by (i) generating test cases that modify the dataset by introducing bias therein and (ii) running the bias checker against the modified dataset; testing the bias checking policy of the bias checker by generating a plurality of test cases and running the bias checker against the plurality of test cases; and providing a notification to a user regarding whether the bias checker failed to indicate bias for one or more of the plurality of attributes.

Abstract: In an embodiment, a method for inspecting and transforming a machine learning model includes receiving a request that includes the machine learning model and a configuration object that provides an indication of a selected strategy. In the embodiment, the method includes creating a partially specified task graph that includes a first placeholder node for a future expanded task node. In the embodiment, the method includes performing a dynamic expansion and execution phase that includes, repeatedly (a) using a cognitive engine to evaluate whether to revise the partially specified task graph based at least in part on the selected strategy, and (b) using a processor-based execution engine to perform an action specified by the complete node. In an embodiment, the dynamic expansion and execution phase repeats until after the cognitive engine adds a consolidated results node.

Abstract: Methods, systems, and computer program products for detecting contextual bias in text are provided herein. A computer-implemented method includes identifying, by a machine learning network, a protected attribute in one or more data samples; processing the identified data samples using a first sub-network of the machine learning network, wherein the first sub-network is configured to determine a plurality of contexts of the protected attribute across the identified data samples; determining an impact of each of the plurality of contexts on a second sub-network of the machine learning network, wherein the second sub-network of the machine learning network is configured to classify a given data sample into one of a plurality of classes; and adjusting the second sub-network of the machine learning to account for the impact of at least one of the plurality of contexts on the second sub-network.

Abstract: Methods, systems, and computer program products for automatically modifying responses from generative models using artificial intelligence techniques are provided herein. A computer-implemented method includes obtaining data pertaining to at least one conversation involving at least one automated conversation exchange software program and at least one user; identifying, among words proposed by the at least one automated conversation exchange software program in connection with the at least one conversation, words qualifying as belonging to one or more predetermined categories by processing the obtained data using artificial intelligence techniques; determining, by processing the identified words and at least one word-based data source, one or more alternate words; modifying at least a portion of the proposed words by replacing at least a portion of the identified words with at least a portion of the one or more alternate words; and performing at least one automated action based on the modifying.

Abstract: Methods, systems, and computer program products for priority-based, accuracy-controlled individual fairness of unstructured text are provided herein. A method includes identifying one or more samples in a set of data used to train a machine learning model having at least one attribute; generating counterfactual samples for each of the one or more identified samples; calculating scores for the one or more identified samples based at least in part on output of the machine learning model with respect to the counterfactual samples, wherein the scores indicate a relative level of bias between the one or more identified samples corresponding to the at least one attribute; creating an enhanced set of data at least in part by supplementing at least a portion of the identified samples with the corresponding counterfactual samples based on the calculated scores; and training the machine learning model using the enhanced set of data.

Abstract: The exemplary embodiments disclose a method, a computer program product, and a computer system for determining that one or more model pipelines satisfy one or more constraints. The exemplary embodiments may include detecting a user uploading data, one or more constraints, and one or more model pipelines, collecting the data, the one or more constraints, and the one or more model pipelines, and determining that one or more of the model pipelines satisfies all of the one or more constraints based on applying one or more algorithms to the collected data, constraints, and model pipelines.

Abstract: In an embodiment, a method for inspecting and transforming a machine learning model includes receiving a request that includes the machine learning model and a configuration object that provides an indication of a selected strategy. In the embodiment, the method includes creating a partially specified task graph that includes a first placeholder node for a future expanded task node. In the embodiment, the method includes performing a dynamic expansion and execution phase that includes, repeatedly (a) using a cognitive engine to evaluate whether to revise the partially specified task graph based at least in part on the selected strategy, and (b) using a processor-based execution engine to perform an action specified by the complete node. In an embodiment, the dynamic expansion and execution phase repeats until after the cognitive engine adds a consolidated results node.

Abstract: Methods, systems, and computer program products for active explanation guided learning are provided herein. A computer-implemented method includes identifying a subset of training examples, from a set of training examples, based on at least one of (i) an uncertainty metric computed for each one of the training examples and (ii) an influence metric computed for each one of the training examples; outputting said subset of training examples to a user; obtaining, from the user, a user explanation for each training example in said subset of training examples, wherein each of the user explanations identifies at least one part of the corresponding training example; and training a machine learning model based at least in part on the user explanations, wherein said training comprises prioritizing the identified parts of the training examples in the subset.

Abstract: One embodiment provides a method, including: receiving a dataset and a model corresponding to a bias checker, wherein the bias checker detects bias within both the dataset and the model, based upon a bias checking algorithm and a bias checking policy, wherein the dataset comprises a plurality of attributes; testing the bias checking algorithm of the bias checker by (i) generating test cases that modify the dataset by introducing bias therein and (ii) running the bias checker against the modified dataset; testing the bias checking policy of the bias checker by generating a plurality of test cases and running the bias checker against the plurality of test cases; and providing a notification to a user regarding whether the bias checker failed to indicate bias for one or more of the plurality of attributes.

Abstract: One embodiment provides a method, including: deploying a machine learning model, wherein the deployed machine learning model is used in responding to queries from users; receiving, at the deployed machine learning model, input from a user; identifying a type of machine learning model attack corresponding to the received input; computing, responsive to receiving the input, a resiliency score of the machine learning model, wherein the resiliency score indicates resistance of the machine learning model against the identified type of attack; and performing an action responsive to the computed resiliency score.

Abstract: One embodiment provides a method, including: receiving, from an application, information for storage in a storage management system, wherein the storage management system comprises a plurality of storage layers, each storage layer having a different performance and a different cost than the other storage layers; labeling the information with one of a plurality of labels relevant to the application; assigning the information a performance tolerance value based upon the label of the information, wherein the performance tolerance value comprises an estimate of the performance requirement required by the application storing the information; determining a storage layer for storage of the information, wherein the determining comprises identifying one of the plurality of storage layers corresponding to the label of the information and updating metadata of a logical volume corresponding to the information with the performance tolerance value; and sending the information to the determined storage layer.

Abstract: One embodiment provides a method, including: receiving, from an application, information for storage in a storage management system, wherein the storage management system comprises a plurality of storage layers, each storage layer having a different performance and a different cost than the other storage layers; labeling the information with one of a plurality of labels relevant to the application; assigning the information a performance tolerance value based upon the label of the information, wherein the performance tolerance value comprises an estimate of the performance requirement required by the application storing the information; determining a storage layer for storage of the information, wherein the determining comprises identifying one of the plurality of storage layers corresponding to the label of the information and updating metadata of a logical volume corresponding to the information with the performance tolerance value; and sending the information to the determined storage layer.

Abstract: One embodiment provides a method, including: deploying a machine learning model, wherein the machine learning model is used in responding to queries from users; receiving, at the deployed machine learning model, input from at least one entity; determining that the at least one entity is an adversary attempting to retrain and/or steal the deployed machine learning model; and providing, in view of the determining that the at least one entity is an adversary, an altered response, wherein the altered response comprises at least one of: a response from a machine learning model other than the deployed machine learning model and a response from the deployed machine learning model altered with errors.

Abstract: One embodiment provides a method, including: deploying a machine learning model, wherein the deployed machine learning model is used in responding to queries from users; receiving, at the deployed machine learning model, input from a user; identifying a type of machine learning model attack corresponding to the received input; computing, responsive to receiving the input, a resiliency score of the machine learning model, wherein the resiliency score indicates resistance of the machine learning model against the identified type of attack; and performing an action responsive to the computed resiliency score.