Irina Rish
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Irina Rish has a total of 64 co-authors. Here are the top co-authors by number of shared publications:
- Guillermo Cecchi
IBM T.J. Watson Research Center, Yorktown Heights, USA
9 shared publications - Aram Galstyan
University of Southern California ♠ University of California Los Angeles ♡ The
3 shared publications - Sahil Garg
USC Information Sciences Institute
3 shared publications Greg Ver Steeg
USC Information Sciences Institute
2 shared publicationsGuillermo A. Cecchi
IBM T.J. Watson Research Center
2 shared publications- Pablo Polosecki
IBM T. J. Watson Research Center, Yorktown Heights, New York City, NY, USA
2 shared publications
Latest Publications
- Open Access
Contextual Bandit with Adaptive Feature Extraction
We consider an online decision making setting known as contextual bandit problem, and propose an approach for improving contextual bandit performance by using an adaptive feature extraction (representation learning) based on online clustering. Our approach starts with an off-line pre-training on unlabeled history of contexts (which can be exploited by our approach, but not by the standard contextual bandit), followed by an online selection and adaptation of encoders. Specifically, given an input sample (context), the proposed approach selects the most appropriate encoding function to extract a feature vector which becomes an input for a contextual bandit, and updates both the bandit and the encoding function based on the context and on the feedback (reward). Our experiments on a variety of datasets, and both in stationary and non-stationary environments of several kinds demonstrate clear advantages of the proposed adaptive representation learning over the standard contextual bandit based on "raw" input contexts.
Open AccessContextual Bandit with Adaptive Feature Extraction
We consider an online decision making setting known as contextual bandit problem, and propose an approach for improving contextual bandit performance by using an adaptive feature extraction (representation learning) based on online clustering. Our approach starts with an off-line pre-training on unlabeled history of contexts (which can be exploited by our approach, but not by the standard contextual bandit), followed by an online selection and adaptation of encoders. Specifically, given an input sample (context), the proposed approach selects the most appropriate encoding function to extract a feature vector which becomes an input for a contextual bandit, and updates both the bandit and the encoding function based on the context and on the feedback (reward). Our experiments on a variety of datasets, and both in stationary and non-stationary environments of several kinds demonstrate clear advantages of the proposed adaptive representation learning over the standard contextual bandit based on "raw" input contexts.
- Open Access
Modeling Psychotherapy Dialogues with Kernelized Hashcode Representations: A Nonparametric Information-Theoretic Approach
We propose a novel dialogue modeling framework, the first-ever nonparametric kernel functions based approach for dialogue modeling, which learns kernelized hashcodes as compressed text representations; unlike traditional deep learning models, it handles well relatively small datasets, while also scaling to large ones. We also derive a novel lower bound on mutual information, used as a model-selection criterion favoring representations with better alignment between the utterances of participants in a collaborative dialogue setting, as well as higher predictability of the generated responses. As demonstrated on three real-life datasets, including prominently psychotherapy sessions, the proposed approach significantly outperforms several state-of-art neural network based dialogue systems, both in terms of computational efficiency, reducing training time from days or weeks to hours, and the response quality, achieving an order of magnitude improvement over competitors in frequency of being chosen as the best model by human evaluators.
Open AccessModeling Psychotherapy Dialogues with Kernelized Hashcode Representations: A Nonparametric Information-Theoretic Approach
We propose a novel dialogue modeling framework, the first-ever nonparametric kernel functions based approach for dialogue modeling, which learns kernelized hashcodes as compressed text representations; unlike traditional deep learning models, it handles well relatively small datasets, while also scaling to large ones. We also derive a novel lower bound on mutual information, used as a model-selection criterion favoring representations with better alignment between the utterances of participants in a collaborative dialogue setting, as well as higher predictability of the generated responses. As demonstrated on three real-life datasets, including prominently psychotherapy sessions, the proposed approach significantly outperforms several state-of-art neural network based dialogue systems, both in terms of computational efficiency, reducing training time from days or weeks to hours, and the response quality, achieving an order of magnitude improvement over competitors in frequency of being chosen as the best model by human evaluators.
- Open Access
- 17 July 2019
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 33 (01) , 6431-6440
Kernelized Hashcode Representations for Relation Extraction
Kernel methods have produced state-of-the-art results for a number of NLP tasks such as relation extraction, but suffer from poor scalability due to the high cost of computing kernel similarities between natural language structures. A recently proposed technique, kernelized locality-sensitive hashing (KLSH), can significantly reduce the computational cost, but is only applicable to classifiers operating on kNN graphs. Here we propose to use random subspaces of KLSH codes for efficiently constructing an explicit representation of NLP structures suitable for general classification methods. Further, we propose an approach for optimizing the KLSH model for classification problems by maximizing an approximation of mutual information between the KLSH codes (feature vectors) and the class labels. We evaluate the proposed approach on biomedical relation extraction datasets, and observe significant and robust improvements in accuracy w.r.t. state-ofthe-art classifiers, along with drastic (orders-of-magnitude) speedup compared to conventional kernel methods.
- 17 July 2019
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 33 (01) , 6431-6440
Open AccessKernelized Hashcode Representations for Relation Extraction
Kernel methods have produced state-of-the-art results for a number of NLP tasks such as relation extraction, but suffer from poor scalability due to the high cost of computing kernel similarities between natural language structures. A recently proposed technique, kernelized locality-sensitive hashing (KLSH), can significantly reduce the computational cost, but is only applicable to classifiers operating on kNN graphs. Here we propose to use random subspaces of KLSH codes for efficiently constructing an explicit representation of NLP structures suitable for general classification methods. Further, we propose an approach for optimizing the KLSH model for classification problems by maximizing an approximation of mutual information between the KLSH codes (feature vectors) and the class labels. We evaluate the proposed approach on biomedical relation extraction datasets, and observe significant and robust improvements in accuracy w.r.t. state-ofthe-art classifiers, along with drastic (orders-of-magnitude) speedup compared to conventional kernel methods.
- Open Access
Kernelized Hashcode Representations for Relation Extraction
Kernel methods have produced state-of-the-art results for a number of NLP tasks such as relation extraction, but suffer from poor scalability due to the high cost of computing kernel similarities between natural language structures. A recently proposed technique, kernelized locality-sensitive hashing (KLSH), can significantly reduce the computational cost, but is only applicable to classifiers operating on kNN graphs. Here we propose to use random subspaces of KLSH codes for efficiently constructing an explicit representation of NLP structures suitable for general classification methods. Further, we propose an approach for optimizing the KLSH model for classification problems by maximizing an approximation of mutual information between the KLSH codes (feature vectors) and the class labels. We evaluate the proposed approach on biomedical relation extraction datasets, and observe significant and robust improvements in accuracy w.r.t. state-of-the-art classifiers, along with drastic (orders-of-magnitude) speedup compared to conventional kernel methods.
Open AccessKernelized Hashcode Representations for Relation Extraction
Kernel methods have produced state-of-the-art results for a number of NLP tasks such as relation extraction, but suffer from poor scalability due to the high cost of computing kernel similarities between natural language structures. A recently proposed technique, kernelized locality-sensitive hashing (KLSH), can significantly reduce the computational cost, but is only applicable to classifiers operating on kNN graphs. Here we propose to use random subspaces of KLSH codes for efficiently constructing an explicit representation of NLP structures suitable for general classification methods. Further, we propose an approach for optimizing the KLSH model for classification problems by maximizing an approximation of mutual information between the KLSH codes (feature vectors) and the class labels. We evaluate the proposed approach on biomedical relation extraction datasets, and observe significant and robust improvements in accuracy w.r.t. state-of-the-art classifiers, along with drastic (orders-of-magnitude) speedup compared to conventional kernel methods.
- Open Access
Continual Learning with Self-Organizing Maps
Despite remarkable successes achieved by modern neural networks in a wide range of applications, these networks perform best in domain-specific stationary environments where they are trained only once on large-scale controlled data repositories. When exposed to non-stationary learning environments, current neural networks tend to forget what they had previously learned, a phenomena known as catastrophic forgetting. Most previous approaches to this problem rely on memory replay buffers which store samples from previously learned tasks, and use them to regularize the learning on new ones. This approach suffers from the important disadvantage of not scaling well to real-life problems in which the memory requirements become enormous. We propose a memoryless method that combines standard supervised neural networks with self-organizing maps to solve the continual learning problem. The role of the self-organizing map is to adaptively cluster the inputs into appropriate task contexts - without explicit labels - and allocate network resources accordingly. Thus, it selectively routes the inputs in accord with previous experience, ensuring that past learning is maintained and does not interfere with current learning. Out method is intuitive, memoryless, and performs on par with current state-of-the-art approaches on standard benchmarks.
Open AccessContinual Learning with Self-Organizing Maps
Despite remarkable successes achieved by modern neural networks in a wide range of applications, these networks perform best in domain-specific stationary environments where they are trained only once on large-scale controlled data repositories. When exposed to non-stationary learning environments, current neural networks tend to forget what they had previously learned, a phenomena known as catastrophic forgetting. Most previous approaches to this problem rely on memory replay buffers which store samples from previously learned tasks, and use them to regularize the learning on new ones. This approach suffers from the important disadvantage of not scaling well to real-life problems in which the memory requirements become enormous. We propose a memoryless method that combines standard supervised neural networks with self-organizing maps to solve the continual learning problem. The role of the self-organizing map is to adaptively cluster the inputs into appropriate task contexts - without explicit labels - and allocate network resources accordingly. Thus, it selectively routes the inputs in accord with previous experience, ensuring that past learning is maintained and does not interfere with current learning. Out method is intuitive, memoryless, and performs on par with current state-of-the-art approaches on standard benchmarks.
- Open Access
Learning Nonlinear Brain Dynamics: van der Pol Meets LSTM
Many real-world data sets, especially in biology, are produced by highly multivariate and nonlinear complex dynamical systems. In this paper, we focus on brain imaging data, including both calcium imaging and functional MRI data. Standard vector-autoregressive models are limited by their linearity assumptions, while nonlinear general-purpose, large-scale temporal models, such as LSTM networks, typically require large amounts of training data, not always readily available in biological applications; furthermore, such models have limited interpretability. We introduce here a novel approach for learning a nonlinear differential equation model aimed at capturing brain dynamics. Specifically, we propose a variable-projection optimization approach to estimate the parameters of the multivariate (coupled) van der Pol oscillator, and demonstrate that such a model can accurately represent nonlinear dynamics of the brain data. Furthermore, in order to improve the predictive accuracy when forecasting future brain-activity time series, we use this analytical model as an unlimited source of simulated data for pretraining LSTM; such model-specific data augmentation approach consistently improves LSTM performance on both calcium and fMRI imaging data.
Open AccessLearning Nonlinear Brain Dynamics: van der Pol Meets LSTM
Many real-world data sets, especially in biology, are produced by highly multivariate and nonlinear complex dynamical systems. In this paper, we focus on brain imaging data, including both calcium imaging and functional MRI data. Standard vector-autoregressive models are limited by their linearity assumptions, while nonlinear general-purpose, large-scale temporal models, such as LSTM networks, typically require large amounts of training data, not always readily available in biological applications; furthermore, such models have limited interpretability. We introduce here a novel approach for learning a nonlinear differential equation model aimed at capturing brain dynamics. Specifically, we propose a variable-projection optimization approach to estimate the parameters of the multivariate (coupled) van der Pol oscillator, and demonstrate that such a model can accurately represent nonlinear dynamics of the brain data. Furthermore, in order to improve the predictive accuracy when forecasting future brain-activity time series, we use this analytical model as an unlimited source of simulated data for pretraining LSTM; such model-specific data augmentation approach consistently improves LSTM performance on both calcium and fMRI imaging data.
- Open Access
Learning Neural Markers of Schizophrenia Disorder Using Recurrent Neural Networks
Smart systems that can accurately diagnose patients with mental disorders and identify effective treatments based on brain functional imaging data are of great applicability and are gaining much attention. Most previous machine learning studies use hand-designed features, such as functional connectivity, which does not maintain the potential useful information in the spatial relationship between brain regions and the temporal profile of the signal in each region. Here we propose a new method based on recurrent-convolutional neural networks to automatically learn useful representations from segments of 4-D fMRI recordings. Our goal is to exploit both spatial and temporal information in the functional MRI movie (at the whole-brain voxel level) for identifying patients with schizophrenia.
Open AccessLearning Neural Markers of Schizophrenia Disorder Using Recurrent Neural Networks
Smart systems that can accurately diagnose patients with mental disorders and identify effective treatments based on brain functional imaging data are of great applicability and are gaining much attention. Most previous machine learning studies use hand-designed features, such as functional connectivity, which does not maintain the potential useful information in the spatial relationship between brain regions and the temporal profile of the signal in each region. Here we propose a new method based on recurrent-convolutional neural networks to automatically learn useful representations from segments of 4-D fMRI recordings. Our goal is to exploit both spatial and temporal information in the functional MRI movie (at the whole-brain voxel level) for identifying patients with schizophrenia.
- 29 August 2017
- Published by Springer Nature
- Vol. 1613, 479-504
Functional Network Disruptions in Schizophrenia
It has been long recognized that schizophrenia, unlike certain other mental disorders, appears to be delocalized, i.e., difficult to attribute to a dysfunction of a few specific brain areas, and may be better understood as a disruption of brain’s emergent network properties. In this chapter, we focus on topological properties of functional brain networks obtained from fMRI data, and demonstrate that some of those properties can be used as discriminative features of schizophrenia in multivariate predictive setting. While the prior work on schizophrenia networks has been primarily focused on discovering statistically significant differences in network properties, this work extends the prior art by exploring the generalization (prediction) ability of network models for schizophrenia, which is not necessarily captured by such significance tests. Moreover, we show that significant disruption of the topological and spatial structure of functional MRI networks in schizophrenia (a) cannot be explained by a disruption to area-based task-dependent responses, i.e., indeed relates to the emergent properties, (b) is global in nature, affecting most dramatically long-distance correlations, and (c) can be leveraged to achieve high classification accuracy (93%) when discriminating between schizophrenic vs. control subjects based just on a single fMRI experiment using a simple auditory task.
- 29 August 2017
- Published by Springer Nature
- Vol. 1613, 479-504
Functional Network Disruptions in Schizophrenia
It has been long recognized that schizophrenia, unlike certain other mental disorders, appears to be delocalized, i.e., difficult to attribute to a dysfunction of a few specific brain areas, and may be better understood as a disruption of brain’s emergent network properties. In this chapter, we focus on topological properties of functional brain networks obtained from fMRI data, and demonstrate that some of those properties can be used as discriminative features of schizophrenia in multivariate predictive setting. While the prior work on schizophrenia networks has been primarily focused on discovering statistically significant differences in network properties, this work extends the prior art by exploring the generalization (prediction) ability of network models for schizophrenia, which is not necessarily captured by such significance tests. Moreover, we show that significant disruption of the topological and spatial structure of functional MRI networks in schizophrenia (a) cannot be explained by a disruption to area-based task-dependent responses, i.e., indeed relates to the emergent properties, (b) is global in nature, affecting most dramatically long-distance correlations, and (c) can be leveraged to achieve high classification accuracy (93%) when discriminating between schizophrenic vs. control subjects based just on a single fMRI experiment using a simple auditory task.
- Open Access
- 16 May 2017
- Published by Springer Nature in Schizophrenia
- Vol. 3 (1) , 22
Learning stable and predictive network-based patterns of schizophrenia and its clinical symptoms
Schizophrenia is often associated with disrupted brain connectivity. However, identifying specific neuroimaging-based patterns pathognomonic for schizophrenia and related symptom severity remains a challenging open problem requiring large-scale data-driven analyses emphasizing not only statistical significance but also stability across multiple datasets, contexts and cohorts. Accurate prediction on previously unseen subjects, or generalization, is also essential for any useful biomarker of schizophrenia. In order to build a predictive model based on functional network feature patterns, we studied whole-brain fMRI functional networks, both at the voxel level and lower-resolution supervoxel level. Targeting Auditory Oddball task data on the FBIRN fMRI dataset (n = 95), we considered node-degree and link-weight network features and evaluated stability and generalization accuracy of statistically significant feature sets in discriminating patients vs. controls. We also applied sparse multivariate regression (elastic net) to whole-brain functional connectivity features, for the first time, to derive stable predictive features for symptom severity. Whole-brain link-weight features achieved 74% accuracy in identifying patients and were more stable than voxel-wise node-degrees. Link-weight features predicted severity of several negative and positive symptom scales, including inattentiveness and bizarre behavior. The most-significant, stable and discriminative functional connectivity changes involved increased correlations between thalamus and primary motor/primary sensory cortex, and between precuneus (BA7) and thalamus, putamen, and Brodmann areas BA9 and BA44. Precuneus, along with BA6 and primary sensory cortex, was also involved in predicting severity of several symptoms. Overall, the proposed multi-step methodology may help identify more reliable multivariate patterns allowing for accurate prediction of schizophrenia and its symptoms severity.
- 16 May 2017
- Published by Springer Nature in Schizophrenia
- Vol. 3 (1) , 22
Open AccessLearning stable and predictive network-based patterns of schizophrenia and its clinical symptoms
Schizophrenia is often associated with disrupted brain connectivity. However, identifying specific neuroimaging-based patterns pathognomonic for schizophrenia and related symptom severity remains a challenging open problem requiring large-scale data-driven analyses emphasizing not only statistical significance but also stability across multiple datasets, contexts and cohorts. Accurate prediction on previously unseen subjects, or generalization, is also essential for any useful biomarker of schizophrenia. In order to build a predictive model based on functional network feature patterns, we studied whole-brain fMRI functional networks, both at the voxel level and lower-resolution supervoxel level. Targeting Auditory Oddball task data on the FBIRN fMRI dataset (n = 95), we considered node-degree and link-weight network features and evaluated stability and generalization accuracy of statistically significant feature sets in discriminating patients vs. controls. We also applied sparse multivariate regression (elastic net) to whole-brain functional connectivity features, for the first time, to derive stable predictive features for symptom severity. Whole-brain link-weight features achieved 74% accuracy in identifying patients and were more stable than voxel-wise node-degrees. Link-weight features predicted severity of several negative and positive symptom scales, including inattentiveness and bizarre behavior. The most-significant, stable and discriminative functional connectivity changes involved increased correlations between thalamus and primary motor/primary sensory cortex, and between precuneus (BA7) and thalamus, putamen, and Brodmann areas BA9 and BA44. Precuneus, along with BA6 and primary sensory cortex, was also involved in predicting severity of several symptoms. Overall, the proposed multi-step methodology may help identify more reliable multivariate patterns allowing for accurate prediction of schizophrenia and its symptoms severity.
- 1 March 2017
- Published by IBM in IBM Journal of Research and Development
- Vol. 61 (2/3) , 3:1-3:9
Holographic brain: Distributed versus local activation patterns in fMRI
- 1 March 2017
- Published by IBM in IBM Journal of Research and Development
- Vol. 61 (2/3) , 3:1-3:9
Holographic brain: Distributed versus local activation patterns in fMRI