Abstract: Techniques are described for training and/or utilizing sub-agent machine learning models to generate candidate dialog responses. In various implementations, a user-facing dialog agent (202, 302), or another component on its behalf, selects one of the candidate responses which is closest to user defined global priority objectives (318). Global priority objectives can include values (306) for a variety of dialog features such as emotion, confusion, objective-relatedness, personality, verbosity, etc. In various implementations, each machine learning model includes an encoder portion and a decoder portion. Each encoder portion and decoder portion can be a recurrent neural network (RNN) model, such as a RNN model that includes at least one memory layer, such as a long short-term memory (LSTM) layer.

Abstract: Techniques are described for training and/or utilizing sub-agent machine learning models to generate candidate dialog responses. In various implementations, a user-facing dialog agent (202, 302), or another component on its behalf, selects one of the candidate responses which is closest to user defined global priority objectives (318). Global priority objectives can include values (306) for a variety of dialog features such as emotion, confusion, objective-relatedness, personality, verbosity, etc. In various implementations, each machine learning model includes an encoder portion and a decoder portion. Each encoder portion and decoder portion can be a recurrent neural network (RNN) model, such as a RNN model that includes at least one memory layer, such as a long short-term memory (LSTM) layer.

Abstract: Techniques are described herein for training and applying memory neural networks, such as “condensed” memory neural networks (“C-MemNN”) and/or “average” memory neural networks (“A-MemNN”). In various embodiments, the memory neural networks may be iteratively trained using training data in the form of free form clinical notes and clinical reference documents. In various embodiments, during each iteration of the training, a so-called “condensed” memory state may be generated and used as part of the next iteration. Once trained, a free form clinical note associated with a patient may be applied as input across the memory neural network to predict one or more diagnoses or outcomes of the patient.

Abstract: The described embodiments relate to systems, methods, and apparatus for providing a multimodal deep memory network (200) capable of generating patient diagnoses (222). The multimodal deep memory network can employ different neural networks, such as a recurrent neural network and a convolution neural network, for creating embeddings (204, 214, 216) from medical images (212) and electronic health records (206). Connections between the input embeddings (204) and diagnoses embeddings (222) can be based on an amount of attention that was given to the images and electronic health records when creating a particular diagnosis. For instance, the amount of attention can be characterized by data (110) that is generated based on sensors that monitor eye movements of clinicians observing the medical images and electronic health records.

Abstract: Techniques are described herein for training and applying memory neural networks, such as “condensed” memory neural networks (“C-MemNN”) and/or “average” memory neural networks (“A-MemNN”). In various embodiments, the memory neural networks may be iteratively trained using training data in the form of free form clinical notes and clinical reference documents. In various embodiments, during each iteration of the training, a so-called “condensed” memory state may be generated and used as part of the next iteration. Once trained, a free form clinical note associated with a patient may be applied as input across the memory neural network to predict one or more diagnoses or outcomes of the patient.

Abstract: A system and method for unsupervised training of a text report identification machine learning model, including: labeling a first set of unlabeled text reports using a seed dictionary to identify concepts in the unlabeled text reports; inputting images associated with the first set of seed-labeled text reports into an auto-encoder to obtain an encoded first set of images; calculating a set of first correlation matrices as a dot product of the first encoded images with their associated text report features; determining a distance between the set of first correlation matrices and a filter bank value associated with the auto-encoder; identifying a first set of validated images as the images in the first set of images that have a distance less than a threshold value; and training the text report machine learning model using the labeled text reports associated with the set of first validated images.

Abstract: Techniques described herein relate to semi-supervised training and application of stacked autoencoders and other classifiers for predictive and other purposes. In various embodiments, a semi-supervised model (108) may be trained for sentence classification, and may combine what is referred to herein as a “residual stacked de-noising autoencoder” (“RSDA”) (220), which may be unsupervised, with a supervised classifier (218) such as a classification neural network (e.g., a multilayer perceptron, or “MLP”). In various embodiments, the RSDA may be a stacked denoising autoencoder that may or may not include one or more residual connections. If present, the residual connections may help the RSDA “remember” forgotten information across multiple layers. In various embodiments, the semi-supervised model may be trained with unlabeled data (for the RSDA) and labeled data (for the classifier) simultaneously.

Abstract: Techniques are described for training and/or utilizing sub-agent machine learning models to generate candidate dialog responses. In various implementations, a user-facing dialog agent (202, 302), or another component on its behalf, selects one of the candidate responses which is closest to user defined global priority objectives (318). Global priority objectives can include values (306) for a variety of dialog features such as emotion, confusion, objective-relatedness, personality, verbosity, etc. In various implementations, each machine learning model includes an encoder portion and a decoder portion. Each encoder portion and decoder portion can be a recurrent neural network (RNN) model, such as a RNN model that includes at least one memory layer, such as a long short-term memory (LSTM) layer.

Abstract: The present disclosure pertains to a paraphrase generation system. The system comprises one or more hardware processors and/or other components. The system is configured to obtain a training corpus. The training corpus comprises language and known paraphrases of the language. The system is configured to generate, based on the training corpus, a word-level attention-based model and a character-level attention-based model. The system is configured to provide one or more candidate paraphrases of a natural language input based on both the word-level and character-level attention-based models. The word-level attention-based model is a word-level bidirectional long short term memory (LSTM) network and the character-level attention-based model is a character-level bidirectional LSTM network. The word-level and character level LSTM networks are generated based on words and characters in the training corpus.

Abstract: Techniques are described herein for training machine learning models to simplify (e.g., paraphrase) complex textual content by ensuring that the machine learning models jointly learn both semantic alignment and notions of simplicity. In various embodiments, an input textual segment having multiple tokens and being associated with a first measure of simplicity may be applied as input across a trained machine learning model to generate an output textual segment. The output textual segment may be is semantically aligned with the input textual segment and associated with a second measure of simplicity that is greater than the first measure of simplicity (e.g., a paraphrase thereof). The trained machine learning model may include an encoder portion and a decoder portion, as well as control layer(s) trained to maximize the second measure of simplicity by replacing token(s) of the input textual segment with replacement token(s).

Abstract: Methods and systems for interacting with a user. Systems in accordance with various embodiments described herein provide a collection of models that are each trained to perform a specific function. These models may be categorized into static models that are trained on an existing corpus of information and dynamic models that are trained based on real-time interactions with users. Collectively, the models provide appropriate communications for a user.

Abstract: A method of producing an unsupervised constrained text simplification autoencoder including an encoder and a constrained decoder, including: encoding, by the encoder, input text to produce a code; combining a complexity parameter with the code; decoding, by constrained decoder, the combined code to produce a plurality of outputs, wherein the constrained decoder uses a dropout function to randomize the parameters of the constrained decoder; evaluating a loss function for each of the plurality of outputs, wherein the loss function is based upon the complexity parameter, indicates an achieved text simplification level, and produces an output indicating the difference between the achieved text simplification level and a desired text simplification level; and optimizing the constrained text simplification autoencoder by repeatedly evaluating the loss function for each input text in an input text training data set while varying parameters of the encoder, the parameters of the constrained decoder, and the complexity

Abstract: Techniques are described herein for training machine learning models to simplify (e.g., paraphrase) complex textual content by ensuring that the machine learning models jointly learn both semantic alignment and notions of simplicity. In various embodiments, an input textual segment having multiple tokens and being associated with a first measure of simplicity may be applied as input across a trained machine learning model to generate an output textual segment. The output textual segment may be is semantically aligned with the input textual segment and associated with a second measure of simplicity that is greater than the first measure of simplicity (e.g., a paraphrase thereof). The trained machine learning model may include an encoder portion and a decoder portion, as well as control layer(s) trained to maximize the second measure of simplicity by replacing token(s) of the input textual segment with replacement token(s).

Abstract: The described embodiments relate to systems, methods, and apparatus for providing a multimodal deep memory network (200) capable of generating patient diagnoses (222). The multimodal deep memory network can employ different neural networks, such as a recurrent neural network and a convolution neural network, for creating embeddings (204, 214, 216) from medical images (212) and electronic health records (206). Connections between the input embeddings (204) and diagnoses embeddings (222) can be based on an amount of attention that was given to the images and electronic health records when creating a particular diagnosis. For instance, the amount of attention can be characterized by data (110) that is generated based on sensors that monitor eye movements of clinicians observing the medical images and electronic health records.

Abstract: Techniques described herein relate to semi-supervised training and application of stacked autoencoders and other classifiers for predictive and other purposes. In various embodiments, a semi-supervised model (108) may be trained for sentence classification and may combine what is referred to herein as a “residual stacked de-noising autoencoder” (“RSDA”) (220), which may be unsupervised, with a supervised classifier (218) such as a classification neural network (e.g., a multilayer perceptron, or “MLP”). In various embodiments, the RSDA may be a stacked denoising autoencoder that may or may not include one or more residual connections. If present, the residual connections may help the RSDA “remember” forgotten information across multiple layers. In various embodiments, the semi-supervised model may be trained with unlabeled data (for the RSDA) and labeled data (for the classifier) simultaneously.

Abstract: Techniques are described herein for training and applying memory neural networks, such as “condensed” memory neural networks (“C-MemNN”) and/or “average” memory neural networks (“A-MemNN”). In various embodiments, the memory neural networks may be iteratively trained using training data in the form of free form clinical notes and clinical reference documents. In various embodiments, during each iteration of the training, a so-called “condensed” memory state may be generated and used as part of the next iteration. Once trained, a free form clinical note associated with a patient may be applied as input across the memory neural network to predict one or more diagnoses or outcomes of the patient.

Abstract: The present disclosure pertains to a paraphrase generation system. The system comprises one or more hardware processors and/or other components. The system is configured to obtain a training corpus. The training corpus comprises language and known paraphrases of the language. The system is configured to generate, based on the training corpus, a word-level attention-based model and a character-level attention-based model. The system is configured to provide one or more candidate paraphrases of a natural language input based on both the word-level and character-level attention-based models. The word-level attention-based model is a word-level bidirectional long short term memory (LSTM) network and the character-level attention-based model is a character-level bidirectional LSTM network. The word-level and character level LSTM networks are generated based on words and characters in the training corpus.