International Business Machines Corporation

Amit Dhurandhar

Karthikeyan Shanmugam

Ronny Luss

Jayant R Kalagnanam

Guillermo A Cecchi

Kush Raj Varshney

Pin-Yu Chen

Pablo Meyer Rojas

Yada Zhu

Markus Ettl

Robert J Baseman

Stuart A Siegel

Sechan Oh

Marek Petrik

Prasanna Sattigeri

Bruce C Graves

Kartik Ahuja

Maja Vukovic

Ruchir Puri

Pawan Raghunath Chowdhary

Yu Tang

Tsuyoshi Ide

Jiri Navratil

Jun Li Wang

Aleksandra Mojsilovic

Soumyadip Ghosh

Peder Andreas Olsen

Naoki Abe

Amar Prakash Azad

Karthik Sourirajan

Moninder Singh

Saneem Ahmed Chemmengath

Fateh A Tipu

Ioana Monica Baldini Soares

Sholom M Weiss

Brian F White

Rajesh Kumar Ravi

Tejaswini Pedapati

Avinash Balakrishnan

William Stephen Schaefer

Abhishek Kumar

Elkin Dario Gutierrez

Anthony T Mazzatti

Kien T Pham

Gopikrishna Maniachari

A method, computer program product and system are provided to generate perturbed text is provided. A processor receives a string of text from a user. A processor determines one or more classifications for at least one word in the string of text by a classification model. A processor determines a plurality of perturbations of the at least one word based on the one or more classifications, where the plurality of perturbations do not share the same one or more classifications as the at least one word in the string of text. A processor selects a perturbation of the string of text based on (i) an edit distance between the string of text and the plurality of perturbations, and (ii) a fluency metric for each of the plurality of perturbations. A processor provides the perturbation of the string of text to the user.

System and method for generating contrastive explanations for text guided by attributes

A method, computer system, and a computer program product for invariant risk minimization games is provided. The present invention may include defining a plurality of environment-specific classifiers corresponding to a plurality of environments. The present invention may also include constructing an ensemble classifier associated with the plurality of environment-specific classifiers. The present invention may further include initiating a game including a plurality of players corresponding to the plurality of environments. The present invention may also include calculating a nash equilibrium of the initiated game. The present invention may further include determining an ensemble predictor based on the calculated nash equilibrium. The present invention may include deploying the determined ensemble predictor associated with the calculated nash equilibrium to make predictions in a new environment.

Learning robust predictors using game theory

A method, system, and computer program product, including generating, using a linear probe, confidence scores through flattened intermediate representations and theoretically-justified weighting of samples during a training of the simple model using the confidence scores of the intermediate representations.

Simple models using confidence profiles

In an approach to improve the efficiency of solving problem instances by utilizing a machine learning model to solve a sequential optimization problem. Embodiments of the present invention receive a sequential optimization problem for solving and utilize a random initialization to solve a first instance of the sequential optimization problem. Embodiments of the present invention learning, by a computing device a machine learning model, based on a previously stored solution to the first instance of the sequential optimization problem. Additionally, embodiments of the present invention generate, by the machine learning model, one or more subsequent approximate solutions to the sequential optimization problem; and output, by a user interface on the computing device, the one or more subsequent approximate solutions to the sequential optimization problem.

Initializing optimization solvers

In an embodiment, a method for generating contrastive information for a classifier prediction comprises receiving image data representative of an input image, using a deep learning classifier model to predict a first classification for the input image, evaluating the input image using a plurality of classifier functions corresponding to respective high-level features to identify one or more of the high-level features absent from the input image, and identifying, from among the high-level features absent from the input image, a pertinent-negative feature that, if added to the input image, will result in the deep learning classifier model predicting a second classification for the modified input image, the second classification being different from the first classification. In an embodiment, the method includes creating a pertinent-positive image that is a modified version of the input image that has the first classification and fewer than all superpixels of the input image.

Contrastive explanations for images with monotonic attribute functions

A computer-implemented method, system, and non-transitory computer-readable storage medium for enhancing performance of a first model. The first model is trained with a training data set. A second model receives the training data set associated with the first model. The second model provides the first model with a hardness value associated with prediction of each data point of the training data set. The first model determines a confidence value regarding predicting each data point based on the training data set, and determines a ratio of the hardness value of a prediction of each data point by the second model with respect to the confidence value of the first model. The first model is retrained with a re-weighted training data set when the determined ratio is lower than a value of &#x3b2;.

Leveraging simple model predictions for enhancing computational performance

A method, system, and computer program product, including generating a contrastive explanation for a decision of a classifier trained on structured data, highlighting an important feature that justifies the decision, and determining a minimal set of new values for features that alter the decision.

Model agnostic contrastive explanations for structured data

A computer-implemented method, a computer program product, and a computer system for diagnosing anomalies detected by a black-box machine learning model. A computer determines a local variance of a test sample in a test dataset, where the local variance represents uncertainty of a prediction by the black-box machine learning model. The computer initializes optimal compensations for the test sample, where the optimal compensations are optimal perturbations to test sample values of respective components of a multivariate input variable. The computer determines local gradients for the test sample. Based on the local variance and the local gradients, the computer updates the optimal compensations until convergences of the optimal compensations are reached. Using the optimal compensations, the computer diagnoses the anomalies detected by the black-box machine learning model.

Diagnosing anomalies detected by black-box machine learning models

Embodiments of the invention include methods, systems, and computer program products for predicting olfactory perception. A non-limiting example of the method includes receiving a library including a plurality of indexed olfactory descriptors. The method also includes receiving an olfactory target descriptor. The method also includes calculating a coefficient matrix and a perceptual distance between an indexed olfactory descriptor and an olfactory target descriptor. The method also includes generating a perceptual descriptor prediction for the olfactory target.

Prediction of olfactory and taste perception through semantic encoding

A system for compressing data during neural network training, comprising of memory that stores computer executable components; a processor that executes computer executable components stored in the memory, wherein the computer executable components comprise of a compilation component that compiles respective molecular descriptors regarding a first set of molecules; a perception component that learns human perception information related to olfactory perceptions of the first set of molecules, and generates predictions of human olfactory perceptions of a second set of molecules; a fitting component that fits distance predictions from the perception component regarding the second set of molecules against measured correct classifications regarding the second set of molecules; and a vector component that generates a perceptual vector distance between two olfactory targets.

Predicting human discriminability of odor mixtures

Techniques facilitating interpretable rule generation using loss-preserving transformation are provided. In one example, a computer-implemented method can comprise evaluating, by a system operatively coupled to a processor, an input data set that comprises three data categories. The computer-implemented method can also comprise transforming, by the system, the input data set into a transformed data set. The transformed data set can comprise two data categories determined based on the three data categories. Transforming the input data set can comprise determining a first cost associated with the transformed data set is no greater than a second cost associated with the input data set.

Interpretable rule generation using loss preserving transformation

Interpretable rule generation using loss-preserving transformation

A network is crawled using a trained learning model to identify a set of secondary-source documents related to an event. A hub page from the set of secondary-source documents is identified that includes a link predicted to link to a new relevant secondary-source document. The new document is added to the set of secondary-source documents. Information is extracted from the set of secondary-source documents. Feedback is received indicative of a relevancy level for the extracted information as applied to the event. Each document is classified into one or more categories related to the event, based on the extracted information and the received feedback information. A learning model is trained based on the received feedback.

Humanitarian crisis analysis using secondary information gathered by a focused web crawler

Predicting human olfactory perception based on molecular structure is described. Molecular descriptor data indicative of molecular descriptors associated with a group of molecular samples can be obtained. Olfactory perception indicator (OPI) data for a set of OPIs can also be obtained with respect to the molecular samples. A training model can be executed on the molecular descriptor data and the OPI data to yield an output model that correlates molecular attributes with OPIs for a single individual or across an aggregate of individuals. The output model can be used to predict olfactory perception for a particular compound or mixture based on which OPIs are correlated with molecular descriptors of the compound or mixture in the output model. The output model can also be inverted and used to identify molecular descriptors that are correlated with a desired set of OPIs. A molecular construct having the molecular descriptors can then be generated.

Correlating olfactory perception with molecular structure

Growing community of inventors

Yorktown Heights, NY, United States of America

Amit Dhurandhar