Intel Corporation

Adobe, Inc.

Quan Anh Tran

Qing Ma

Valluri Ramana Rao

John Heck

Trung Huu Bui

Li-Peng Wang

Walter Chang

Franck Dernoncourt

Tsung-Kuan Allen Chou

Tuan Manh Lai

Zhe Lin

Johanna M Swan

Robert L Sankman

Dong S Shim

Jianming Zhang

Ebrahim Andideh

Handong Zhao

Doo Soon Kim

Raiyomand F Aspandiar

Ling Liao

Ankur Agrawal

Hung Hai Bui

Susheel G Jadhav

Jim Maveety

Priyanka Dobriyal

Saeed Fathololoumi

Steve Towle

Xihui Liu

Jin Hyung Lee

Nathan Franklin

Rajiv Jain

Raghuram Narayan

Harry H Fujimoto

Amir Pouran Ben Veyseh

Varun Manjunatha

Donald E Adams

Hanan Bar

Robert N Stark

Long Mai

Boaz Weinfeld

Lidan Wang

Pengyue Wen

Kenneth Brown

Kushal Kafle

Joseph S Hayden, Iii

Yevgeny Anoikin

Byong Man Kim

Zhuowan Li

Aditi Mallik

Semeon Altshuler

Joseph Melki

Khalil Mrini

Kenneth M Brown

Wendai Wang

Xiaozhong Wang

Qiuyu Chen

Methods and apparatus to reduce stress on lasers in optical transceivers are disclosed. An apparatus comprising a printed circuit board (PCB) having a first side and a second side opposite the first side; and a first stiffener attached to the first side of the PCB; and a photonic integrated circuit (PIC) attached to the first stiffener. The first stiffener is between the PIC and the PCB. The apparatus also includes a second stiffener attached to the second side of the PCB.

Methods and apparatus to reduce stress on lasers in optical transceivers

Disclosed herein are optical transceivers with multi-laser modules, as well as related optoelectronic assemblies and methods. In some embodiments, an optical transceiver may include: a first laser and a second laser; an optical output path, wherein an output of the first laser is coupled to the optical output path; and switching circuitry to decouple the output of the first laser from the optical output path and to couple an output of the second laser to the optical output path.

Optical transceivers with multi-laser modules

The present disclosure describes a model for large scale color prediction of objects identified in images. Embodiments of the present disclosure include an object detection network, an attention network, and a color classification network. The object detection network generates object features for an object in an image and may include a convolutional neural network (CNN), region proposal network, or a ResNet. The attention network generates an attention vector for the object based on the object features, wherein the attention network takes a query vector based on the object features, and a plurality of key vector and a plurality of value vectors corresponding to a plurality of colors as input. The color classification network generates a color attribute vector based on the attention vector, wherein the color attribute vector indicates a probability of the object including each of the plurality of colors.

Decompositional learning for color attribute prediction

The present disclosure describes systems and methods for dialog processing and information retrieval. Embodiments of the present disclosure provide a dialog system (e.g., a task-oriented dialog system) with adaptive recurrent hopping and dual context encoding to receive and understand a natural language query from a user, manage dialog based on natural language conversation, and generate natural language responses. For example, a memory network can employ a memory recurrent neural net layer and a decision meta network (e.g., a subnet) to determine an adaptive number of memory hops for obtaining readouts from a knowledge base. Further, in some embodiments, a memory network uses a dual context encoder to encode information from original context and canonical context using parallel encoding layers.

Dialog system with adaptive recurrent hopping and dual context encoding

Systems and methods for natural language processing are described. One or more embodiments of the present disclosure receive a document comprising a plurality of words organized into a plurality of sentences, the words comprising an event trigger word and an argument candidate word, generate word representation vectors for the words, generate a plurality of document structures including a semantic structure for the document based on the word representation vectors, a syntax structure representing dependency relationships between the words, and a discourse structure representing discourse information of the document based on the plurality of sentences, generate a relationship representation vector based on the document structures, and predict a relationship between the event trigger word and the argument candidate word based on the relationship representation vector.

Inducing rich interaction structures between words for document-level event argument extraction

Embodiments disclosed herein include optoelectronic systems and methods of forming such systems. In an embodiment the optoelectronic system comprises a board, and a carrier attached to the board. In an embodiment, a first die is on the carrier. In an embodiment, the first die is a photonics die, and a surface of the first die is covered by an optically transparent layer.

Silicon photonic integrated lens compatible with wafer processing

Certain embodiments involve a method for generating a search result. The method includes processing devices performing operations including receiving a query having a text input by a joint embedding model trained to generate an image result. Training the joint embedding model includes accessing a set of images and textual information. Training further includes encoding the images into image feature vectors based on spatial features. Further, training includes encoding the textual information into textual feature vectors based on semantic information. Training further includes generating a set of image-text pairs based on matches between image feature vectors and textual feature vectors. Further, training includes generating a visual grounding dataset based on spatial information. Training further includes generating a set of visual-semantic joint embeddings by grounding the image-text pairs with the visual grounding dataset. Additionally, operations include generating an image result for display by the joint embedding model based on the text input.

Joint visual-semantic embedding and grounding via multi-task training for image searching

A group captioning system includes computing hardware, software, and/or firmware components in support of the enhanced group captioning contemplated herein. In operation, the system generates a target embedding for a group of target images, as well as a reference embedding for a group of reference images. The system identifies information in-common between the group of target images and the group of reference images and removes the joint information from the target embedding and the reference embedding. The result is a contrastive group embedding that includes a contrastive target embedding and a contrastive reference embedding with which to construct a contrastive group embedding, which is then input to a model to obtain a group caption for the target group of images.

Contrastive captioning for image groups

The present disclosure provides a memory-based neural network for question answering. Embodiments of the disclosure identify meta-evidence nodes in an embedding space, where the meta-evidence nodes represent salient features of a training set. Each element of the training set may include a questions appended to a ground truth answer. The training set may also include questions with wrong answers that are indicated as such. In some examples, a neural Turing machine (NTM) reads a dataset and summarizes the dataset into a few meta-evidence nodes. A subsequent question may be appended to multiple candidate answers to form an input phrase, which may also be embedded in the embedding space. Then, corresponding weights may be identified for each of the meta-evidence nodes. The embedded input phrase and the weighted meta-evidence nodes may be used to identify the most appropriate answer.

Memory-based neural network for question answering

An integrated circuit assembly includes a support (e.g., package substrate or circuit board) and a semiconductor die including a device. The semiconductor die is mounted to the support with the device facing the support. The device can be, for example, a quantum well laser device or a photonics device. A layer of decoupling material is on the device. An underfill material is between the semiconductor die and the support, where the decoupling material is between the device and the underfill material. The decoupling layer decouples stress from transferring from the underfill material into the device. For example, the decoupling material forms only weak bonds with the underfill material and/or a passivation layer on the device, in an embodiment. Weak bonds include non-covalent bonds and non-ionic bonds, for example. The decoupling material can be, for instance, a PTFE film, a poly(p-xylylene) film, a fluorocarbon, or a compound lacking free hydroxyl groups.

Decoupling layer to reduce underfill stress in semiconductor devices

Techniques for training a first neural network (NN) model using a pre-trained second NN model are disclosed. In an example, training data is input to the first and second models. The training data includes masked tokens and unmasked tokens. In response, the first model generates a first prediction associated with a masked token and a second prediction associated with an unmasked token, and the second model generates a third prediction associated with the masked token and a fourth prediction associated with the unmasked token. The first model is trained, based at least in part on the first, second, third, and fourth predictions. In another example, a prediction associated with a masked token, a prediction associated with an unmasked token, and a prediction associated with whether two sentences of training data are adjacent sentences are received from each of the first and second models. The first model is trained using the predictions.

Training of neural network based natural language processing models using dense knowledge distillation

A computer-implemented method is disclosed for determining one or more characteristics of a dialog between a computer system and user. The method may comprise receiving a system utterance comprising one or more tokens defining one or more words generated by the computer system; receiving a user utterance comprising one or more tokens defining one or more words uttered by a user in response to the system utterance, the system utterance and the user utterance forming a dialog context; receiving one or more utterance candidates comprising one or more tokens; for each utterance candidate, generating an input sequence combining the one or more tokens of each of the system utterance, the user utterance, and the utterance candidate; and for each utterance candidate, evaluating the generated input sequence with a model to determine a probability that the utterance candidate is relevant to the dialog context.

Methods and systems for determining characteristics of a dialog between a computer and a user

The technology described herein is directed to a reinforcement learning based framework for training a natural media agent to learn a rendering policy without human supervision or labeled datasets. The reinforcement learning based framework feeds the natural media agent a training dataset to implicitly learn the rendering policy by exploring a canvas and minimizing a loss function. Once trained, the natural media agent can be applied to any reference image to generate a series (or sequence) of continuous-valued primitive graphic actions, e.g., sequence of painting strokes, that when rendered by a synthetic rendering environment on a canvas, reproduce an identical or transformed version of the reference image subject to limitations of an action space and the learned rendering policy.

Semantic image manipulation using visual-semantic joint embeddings

Systems and methods for parsing natural language sentences using an artificial neural network (ANN) are described. Embodiments of the described systems and methods may generate a plurality of word representation matrices for an input sentence, wherein each of the word representation matrices is based on an input matrix of word vectors, a query vector, a matrix of key vectors, and a matrix of value vectors, and wherein a number of the word representation matrices is based on a number of syntactic categories, compress each of the plurality of word representation matrices to produce a plurality of compressed word representation matrices, concatenate the plurality of compressed word representation matrices to produce an output matrix of word vectors, and identify at least one word from the input sentence corresponding to a syntactic category based on the output matrix of word vectors.

Interpretable label-attentive encoder-decoder parser

Growing community of inventors

Fremont, CA, United States of America

Quan Anh Tran