International Business Machines Corporation

Kyndryl, Inc.

Debasisha Padhi

Anuradha Bhamidipaty

Yaoping Ruan

Constantin Mircea Adam

Frederick Y-f Wu

Ashish Kundu

Sreekrishnan Venkateswaran

Vysakh K Chandran

Sai Zeng

Milton H Hernandez

Nikolaos Anerousis

Shubhi Asthana

Fabio Minoru Tanada

Jayan Nallacherry

Lohitashwa Thyagaraj

Richard Jay Cohen

Firas Bouz

Amol Bhaskar Mahamuni

Rahul Chenny

Shashidhar Sastry

Sarin Kumar Thayyilsubramanian

Souri Subudhi

Nidhi Patel

A method is provided that includes configuring a cellular phone to function as a mobile debit card by installing an application for contactless interfacing with ATMs. The method supplements the application with a neural network (NN) based biometric verification process configured to reduce an incorrect user error value over time to increasingly harden the application to undesired intrusion. The NN based biometric verification process comprises: performing an initial face recognition; greeting the user with one of a plurality of questions in a specific language of the user and evaluating a pre-defined answer provided from the user in the specific language; and detecting, in an acoustic utterance having dialogue in support of an ATM session, a voice and a prosody style indicative of the user in combination with lip and face movements made by the user corresponding to and in synchronization with the acoustic utterance.

Secure and contactless debit transactions

A method, system, and computer program product for prioritizing endpoints to be checked during a change window based on certain criteria. The method may include receiving a request for processing from a plurality of servers. The method may also include determining a priority for each server of the plurality of servers based on specified criteria, where the specified criteria includes at least compliance-check history. The method may also include determining whether each server belongs to one or more groups. The method may also include determining a notification order for the plurality of servers based on the priority and whether each server belongs to the one or more groups. The method may also include sending a notification to each server in the notification order.

Prioritize endpoint selection based on criteria

A computer-implemented method in accordance with one embodiment includes, in response to a submission of an input dataset to an artificially intelligent application, receiving an explanation from each module of the application. The modules are configured within the application in a serial sequence in which each module, upon receiving the input dataset and any input generated by an immediately preceding module of the serial sequence, generates output that is forwarded as input to a next module, if any, in the sequence. A determination is made that at least two of the received explanations are semantically inconsistent.

Semantic consistency of explanations in explainable artificial intelligence applications

An actuator to execute on a server may be automatically selected based on risk of failure and damage to the server. Requirement specification and environment parameters may be received. A subset of actuators may be selected based on a risk threshold from an actuator catalog database storing actuator information and actuator risk metadata associated with a plurality of actuators. The actuator risk metadata may be augmented with risk information. A ranked list of the subset of actuators may be generated based on the actuator risk metadata associated with each actuator in the subset. An actuator in the ranked list may be executed on the server.

Action recommendation to reduce server management errors

An explainable artificially intelligent (XAI) application contains an ordered sequence of artificially intelligent software modules. When an input dataset is submitted to the application, each module generates an output dataset and an explanation that represents, as a set of Boolean expressions, reasoning by which each output element was chosen. If any pair of explanations are determined to be semantically inconsistent, and if this determination is confirmed by further determining that an apparent inconsistency was not a correct response to an unexpected characteristic of the input dataset, nonzero inconsistency scores are assigned to inconsistent elements of the pair of explanations. If the application's overall inconsistency score exceeds a threshold value, the system forwards information about the explanation, the offending modules, and the input dataset to a downstream machine-learning component that uses this information to train the application to better respond to future input that shares certain characteristics with the current input.

Aspects of the present invention disclose a method, computer program product, and system for multi-factor authentication. In response to a request for an action, the method includes one or more processors whether a first authentication credential passes validation. In response to determining that the first authentication credential does pass validation, the method further includes one or more processors determining a second authentication credential, wherein the second authentication credential includes an indication of a wireless connection between a first computing device and a second computing device. The method further includes one or more processors determining whether the second authentication credential passes validation. In response to determining that the second authentication credential passes validation, the method further includes one or more processors allowing execution of the requested response.

Multi-factor authentication utilizing device pairing

A method, computer program product, and a system where a processor(s) obtains, from a data source, a list of objects at different layers of a computing system. The processor(s) generates exploration lists from the list (each exploration list with objects for a layer). The processor(s) identifies updated and new data at the layers associated with the objects on the list; the identified data comprises attributes for each layer. The processor(s) applies machine learning algorithm(s) to enrich the data by identifying dependencies between the attributes for each layer as influencers for one or more key performance indicators of the computing system. The processor(s) generates, from the enriched data, a hierarchy matrix. The processor(s) determines, based on the hierarchy matrix that an event associated with one or more computing resources of the computing system will influence a particular key performance indicator.

Multi-layered computing system attribute dependency

Provisioning infrastructure from visual diagrams can include selecting, with a first neural network, regions of interest within an infrastructure architecture diagram that corresponds to an infrastructure offered by an infrastructure provider for supporting a computer application. With a second neural network, infrastructure resources and interconnections among the infrastructure resources can be identified based on objects appearing within the regions of interest. Properties corresponding to the infrastructure resources identified can be determined with a third neural network. With a fourth neural network, an infrastructure architecture specification can be generated, the infrastructure architecture specification specifying the infrastructure resources, corresponding properties of the infrastructure resources, and interconnections among the infrastructure resources. The infrastructure architecture specification can be used to configure networked resources to support the computer application.

Provisioning infrastructure from visual diagrams

A system, method and program product for providing cognitive behavior security control (CBSC). A system is disclosed that includes: a repository having a plurality of challenges each including an interactive graphical task; a user interface for graphically presenting challenges to users; an enrollment system for assigning challenges to users and determining an authentic response for each user; an authentication system that collects an observed response from a user presented with an assigned challenge and determines a security control result based on a closeness of the observed response with the authentic response of the user.

Cognitive behavioral security controls

A computer-implemented agent process running on a first computer automatically intercepts a command issued from the first computer to execute on a target computer prior to invocation of the command on the target computer. A server profile built for an application running on the target computer that supports the command may be retrieved. At least based on the server profile a risk enforcement policy is dynamically constructed. Based on the risk enforcement policy, one or more computer-executable enforcement actions to perform prior to sending the command to the target computer for execution is determined. Based on executing of one or more of the computer-executable enforcement actions, the command may be transmitted to execute on the target computer or prevented from executing on the target computer.

Growing community of inventors

Bengaluru, India

Debasisha Padhi