Abstract: Composite hollow fiber gas separation membranes with improved permeance and separation layer adhesion are manufactured by providing dipping a hollow fiber membrane substrate in a pre-coat layer coating composition followed by drying to thereby provide a pre-coated substrate and dipping the pre-coated substrate in a separation layer coating composition followed by drying to thereby provide the composite hollow fiber gas separation membranes. The pre-coating composition includes a first polymer dissolved in a first solvent and the separation layer composition includes a second polymer dissolved in a second solvent. The first and second polymers are the same or different, each of the first and second polymers is at least 1 wt % soluble in a same third solvent, the first and second solvents are the same or different, the first and third solvents are the same or different, and the second and third solvent are the same or different.

Abstract: The present disclosure provides a multi-factor quantitative analysis method for deformation of a neighborhood tunnel. The method includes the following steps: analyzing monitoring data generated at a tunnel site; simulating collapse occurring at a shallow buried section of a tunnel; determining the degree of influence of each factor on the tunnel and a stratum; and determining quantitative influence of each factor on tunnel deformation. The present disclosure can not only provide an accurate theoretical basis for the construction of the shallow buried section of the small-distance tunnel, but also guarantee safety and cost saving during tunnel construction.

Abstract: A human-computer interaction method and apparatus. Said method may include: receiving information of at least one modality of a user (201); identifying, on the basis of the information of the at least one modality, intention information of the user and user emotional features corresponding to the intention information (202); determining, on the basis of the intention information, reply information to the user (203); selecting, on the basis of the user emotional features, character emotional features to be fed back to the user (204); and generating, on the basis of the character emotional features and the reply information, a broadcast video of an animated character corresponding to the character emotional features (205).

Abstract: A liquid-solid axial moving bed reaction and regeneration apparatus and a solid acid alkylation process by using the liquid-solid axial moving bed reaction and regeneration apparatus.

Abstract: Technologies related to generating diverse electronic summary documents for a webpage are described herein. A sequence of tokens is extracted from the webpage, and the sequence of tokens is provided to several computer-implemented models. The computer-implemented models output respective sets of candidate assets based upon the sequence of tokens, where the candidate assets are potentially included in an electronic summary document for the webpage. Subsequently, a user query is received, and at least one candidate asset from the candidate assets are selected for inclusion in the electronic summary document based upon the query. Thus, different electronic summary documents can be generated for the webpage when different queries are received.

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: A system (1000) for automated question answering, including: semantic space (210) generated from a corpus of questions and answers; a user interface (1030) configured to receive a question; and a processor (1100) comprising: (i) a question decomposition engine (1050) configured to decompose the question into a domain, a keyword, and a focus word; (ii) a question similarity generator (1060) configured to identify one or more questions in a semantic space using the decomposed question; (iii) an answer extraction and ranking engine (1080) configured to: extract, from the semantic space, answers associated with the one or more identified questions; and identify one or more of the extracted answers as a best answer; and (iv) an answer tuning engine (1090) configured to fine-tune the identified best answer using one or more of the domain, keyword, and focus word; wherein the fine-tuned answer is provided to the user via the user interface.

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: Techniques are described herein for drawing conclusions using free form texts and external resources. In various embodiments, free form input data (202) may be segmented (504) into a plurality of input data segments. A first input data segment may be compared (510) with an external resource (304) to identify a first candidate conclusion. A reinforcement learning trained agent (310) may be applied (512) to make a first determination of whether to accept or reject the first candidate conclusion. Similar actions may be performed with a second input data segment to make a second determination of whether to accept or reject a second candidate conclusion. A final conclusion may be presented (522) based on the first and second determinations of the reinforcement learning trained agent with respect to at least the first candidate conclusion and the second candidate conclusion.

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 skid steer loader includes a chassis frame, a cab, a main beam, and a bucket. The chassis frame is assembled with a walking system to form a movable chassis of the skid steer loader, and a power system of the skid steer loader is assembled to a tail portion of the chassis frame. The cab is arranged on a front portion of the chassis frame in an. overturning manner, and the cab is horizontally placed on the chassis frame when a telescopic support rod is in a retracted state and overturns forward when the telescopic support rod is in an extended state. A tail end of the main beam is movably hinged to the tail portion of the chassis frame by a four-bar linkage mechanism, the bucket is assembled to a front end of the main beam.

Abstract: A medical information retrieval system comprises a natural language processing system that processes a vocal user query to identify key words and phrases. These key words and phrases are provided to an inferencing engine that provides a set of knowledge-based inferences from medical knowledge sources, based on these key words and phrases. Thereafter, these knowledge-based inferences are provided to an information retrieval engine that retrieves a corresponding plurality of medical articles based on these knowledge-based inferences, and ranks each with respect to the knowledge-based inferences. A summary engine receives the ranked articles and creates a model based on the topical keywords and candidate sentences found in the highly ranked articles. A paraphrase engine processes the candidate sentences to provide a summary response based on a knowledge-based paraphrase model. An audio output device renders the summary report as the response to the user's original vocal query.

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: Methods and systems for generating text from a haptic-based input. The system may include an interface for receiving a haptic-based input and a processor executing instructions stored on a memory and providing a model. The model is configured to at least receive the haptic-based input and supply a text describing the haptic-based input using the interface.

Abstract: Carbamate-substituted styryl sulfone compounds have a structure of general formula (I). The compounds not only has a significantly enhanced acetylcholinesterase inhibitory activity compared with carbamates as positive drugs, but also has a significantly enhanced therapeutic effect on Parkinson's disease compared with caffeic acid phenethyl ester compounds having similar structures.

Abstract: Techniques are provided for generating and applying a granular attention hierarchical neural network model to classify a document. In various embodiments, data indicative of the document may be obtained (102) and processed (104) into a first layer of two or more layers of a hierarchical network model using a dual granularity attention mechanism to generate first layer output data, wherein the dual granularity attention mechanism weighs some portions of the data indicative of the document more heavily. Some portions of the data indicative of the document are integrated into the hieratical network model during training of the dual granularity attention mechanism. The first layer output data may be processed (106) in the second of two or more layers of the hierarchical network model to generate second layer output data. A classification label can be generated (108) from the second layer output data.

Abstract: An aromatization catalyst and preparation process and use thereof is set forth. The catalyst comprises an inorganic oxide and a modified Ga-ZSM-5 zeolite, which comprises a modified ZSM-5 zeolite with a hierarchical macro-meso-microporosity and gallium deposited in channels of and/or on surfaces of the modified ZSM-5 zeolite. The hierarchical porosity of the modified ZSM-5 zeolite in the catalyst can reduce diffusion resistance of products during the aromatization reaction, thereby retarding carbon depositing rate and substantially improving catalytic activity, aromatic hydrocarbon selectivity, stability and lifetime of the catalyst. When being used in aromatization of propane, the catalyst exhibits a high stability, a lifetime of more than 320 hours, and a selectivity to aromatic hydrocarbons of up to 73.3 wt. %.

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: Methods and systems for generating a question from free text. The system is trained on a corpus of data and receives a tuple consisting of a paragraph (free text), a focused fact, and a question type. The system implements a language model to find the most optimal combination of words to return a question for the paragraph about the focused fact.