scholarly journals Zero-Shot Adaptive Transfer for Conversational Language Understanding

2019 ◽  
Vol 33 ◽  
pp. 6642-6649 ◽  
Author(s):  
Sungjin Lee ◽  
Rahul Jha
Keyword(s):  
Domain Adaptation ◽  
State Of The Art ◽  
Personal Digital Assistant ◽  
Conversational Agents ◽  
Language Understanding ◽  
Training Time ◽  
Large Margin ◽  
Previous State ◽  
Conversational Language ◽  
Transfer Method

Conversational agents such as Alexa and Google Assistant constantly need to increase their language understanding capabilities by adding new domains. A massive amount of labeled data is required for training each new domain. While domain adaptation approaches alleviate the annotation cost, prior approaches suffer from increased training time and suboptimal concept alignments. To tackle this, we introduce a novel Zero-Shot Adaptive Transfer method for slot tagging that utilizes the slot description for transferring reusable concepts across domains, and enjoys efficient training without any explicit concept alignments. Extensive experimentation over a dataset of 10 domains relevant to our commercial personal digital assistant shows that our model outperforms previous state-of-the-art systems by a large margin, and achieves an even higher improvement in the low data regime.

Multi-Resolution Autoregressive Graph-to-Graph Translation for Molecules

2019 ◽  
Author(s):  
Wengong Jin ◽  
Regina Barzilay ◽  
Tommi S Jaakkola
Keyword(s):  
Drug Discovery ◽  
State Of The Art ◽  
Molecular Graph ◽  
Biochemical Properties ◽  
Large Margin ◽  
Previous State ◽  
Translation Methods ◽  
Atom Level ◽  
Precursor Molecules ◽  
Prior State

The problem of accelerating drug discovery relies heavily on automatic tools to optimize precursor molecules to afford them with better biochemical properties. Our work in this paper substantially extends prior state-of-the-art on graph-to-graph translation methods for molecular optimization. In particular, we realize coherent multi-resolution representations by interweaving trees over substructures with the atom-level encoding of the original molecular graph. Moreover, our graph decoder is fully autoregressive, and interleaves each step of adding a new substructure with the process of resolving its connectivity to the emerging molecule. We evaluate our model on multiple molecular optimization tasks and show that our model outperforms previous state-of-the-art baselines by a large margin.

scholarly journals Automatically Paraphrasing via Sentence Reconstruction and Round-trip Translation

2021 ◽  
Author(s):  
Zilu Guo ◽  
Zhongqiang Huang ◽  
Kenny Q. Zhu ◽  
Guandan Chen ◽  
Kaibo Zhang ◽  
...  
Keyword(s):  
Machine Translation ◽  
Question Answering ◽  
Domain Adaptation ◽  
State Of The Art ◽  
Training Data ◽  
Round Trip ◽  
Previous State ◽  
Supervised Methods ◽  
Paraphrase Generation ◽  
Better Than

Paraphrase generation plays key roles in NLP tasks such as question answering, machine translation, and information retrieval. In this paper, we propose a novel framework for paraphrase generation. It simultaneously decodes the output sentence using a pretrained wordset-to-sequence model and a round-trip translation model. We evaluate this framework on Quora, WikiAnswers, MSCOCO and Twitter, and show its advantage over previous state-of-the-art unsupervised methods and distantly-supervised methods by significant margins on all datasets. For Quora and WikiAnswers, our framework even performs better than some strongly supervised methods with domain adaptation. Further, we show that the generated paraphrases can be used to augment the training data for machine translation to achieve substantial improvements.

Multi-Resolution Autoregressive Graph-to-Graph Translation for Molecules

2019 ◽  
Author(s):  
Wengong Jin ◽  
Regina Barzilay ◽  
Tommi S Jaakkola
Keyword(s):  
Drug Discovery ◽  
State Of The Art ◽  
Molecular Graph ◽  
Biochemical Properties ◽  
Large Margin ◽  
Previous State ◽  
Translation Methods ◽  
Atom Level ◽  
Precursor Molecules ◽  
Prior State

The problem of accelerating drug discovery relies heavily on automatic tools to optimize precursor molecules to afford them with better biochemical properties. Our work in this paper substantially extends prior state-of-the-art on graph-to-graph translation methods for molecular optimization. In particular, we realize coherent multi-resolution representations by interweaving trees over substructures with the atom-level encoding of the original molecular graph. Moreover, our graph decoder is fully autoregressive, and interleaves each step of adding a new substructure with the process of resolving its connectivity to the emerging molecule. We evaluate our model on multiple molecular optimization tasks and show that our model outperforms previous state-of-the-art baselines by a large margin.

scholarly journals Alexa Prize — State of the Art in Conversational AI

AI Magazine ◽  
2018 ◽  
Vol 39 (3) ◽  
pp. 40-55 ◽  
Author(s):  
Chandra Khatri ◽  
Anu Venkatesh ◽  
Behnam Hedayatnia ◽  
Raefer Gabriel ◽  
Ashwin Ram ◽  
...  
Keyword(s):  
Natural Language ◽  
State Of The Art ◽  
Natural Language Understanding ◽  
Academic Community ◽  
The State ◽  
First Year ◽  
Grand Challenge ◽  
Conversational Agents ◽  
Language Understanding ◽  
Second Year

To advance the state of the art in conversational AI, Amazon launched the Alexa Prize, a 2.5 million dollar competition that challenges university teams to build conversational agents, or "socialbots", that can converse coherently and engagingly with humans on popular topics for 20 minutes. The Alexa Prize offers the academic community a unique opportunity to perform research at scale with real conversational data obtained by interacting with millions of Alexa users, along with user-provided ratings and feedback, over several months. This enables teams to effectively iterate, improve and evaluate their socialbots throughout the competition. Sixteen teams were selected for the inaugural competition last year. To build their socialbots, the students combined state-of-the-art techniques with their own novel strategies in the areas of Natural Language Understanding and Conversational AI. This article reports on the research conducted over the 2017-2018 year. While the 20 minute grand challenge was not achieved in the first year, the competition produced several conversational agents that advanced the state of the art, are interesting for everyday users to interact with, and help form a baseline for the second year of the competition.

scholarly journals Using spatial-temporal ensembles of convolutional neural networks for lumen segmentation in ureteroscopy

2021 ◽  
Author(s):  
Jorge F. Lazo ◽  
Aldo Marzullo ◽  
Sara Moccia ◽  
Michele Catellani ◽  
Benoit Rosa ◽  
...  
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
State Of The Art ◽  
Automatic Segmentation ◽  
Temporal Information ◽  
Invasive Technique ◽  
Dice Similarity Coefficient ◽  
Specular Reflections ◽  
Lumen Segmentation ◽  
Previous State

Abstract Purpose Ureteroscopy is an efficient endoscopic minimally invasive technique for the diagnosis and treatment of upper tract urothelial carcinoma. During ureteroscopy, the automatic segmentation of the hollow lumen is of primary importance, since it indicates the path that the endoscope should follow. In order to obtain an accurate segmentation of the hollow lumen, this paper presents an automatic method based on convolutional neural networks (CNNs). Methods The proposed method is based on an ensemble of 4 parallel CNNs to simultaneously process single and multi-frame information. Of these, two architectures are taken as core-models, namely U-Net based in residual blocks ($$m_1$$ m 1 ) and Mask-RCNN ($$m_2$$ m 2 ), which are fed with single still-frames I(t). The other two models ($$M_1$$ M 1 , $$M_2$$ M 2 ) are modifications of the former ones consisting on the addition of a stage which makes use of 3D convolutions to process temporal information. $$M_1$$ M 1 , $$M_2$$ M 2 are fed with triplets of frames ($$I(t-1)$$ I ( t - 1 ) , I(t), $$I(t+1)$$ I ( t + 1 ) ) to produce the segmentation for I(t). Results The proposed method was evaluated using a custom dataset of 11 videos (2673 frames) which were collected and manually annotated from 6 patients. We obtain a Dice similarity coefficient of 0.80, outperforming previous state-of-the-art methods. Conclusion The obtained results show that spatial-temporal information can be effectively exploited by the ensemble model to improve hollow lumen segmentation in ureteroscopic images. The method is effective also in the presence of poor visibility, occasional bleeding, or specular reflections.

scholarly journals Bottom-up and Layerwise Domain Adaptation for Pedestrian Detection in Thermal Images

2021 ◽  
Vol 17 (1) ◽  
pp. 1-19
Author(s):  
My Kieu ◽  
Andrew D. Bagdanov ◽  
Marco Bertini
Keyword(s):  
Domain Adaptation ◽  
State Of The Art ◽  
Pedestrian Detection ◽  
Challenging Problem ◽  
Top Down ◽  
Bottom Up ◽  
Security Applications ◽  
Lighting Conditions ◽  
Initial Layers ◽  
Single Modality

Pedestrian detection is a canonical problem for safety and security applications, and it remains a challenging problem due to the highly variable lighting conditions in which pedestrians must be detected. This article investigates several domain adaptation approaches to adapt RGB-trained detectors to the thermal domain. Building on our earlier work on domain adaptation for privacy-preserving pedestrian detection, we conducted an extensive experimental evaluation comparing top-down and bottom-up domain adaptation and also propose two new bottom-up domain adaptation strategies. For top-down domain adaptation, we leverage a detector pre-trained on RGB imagery and efficiently adapt it to perform pedestrian detection in the thermal domain. Our bottom-up domain adaptation approaches include two steps: first, training an adapter segment corresponding to initial layers of the RGB-trained detector adapts to the new input distribution; then, we reconnect the adapter segment to the original RGB-trained detector for final adaptation with a top-down loss. To the best of our knowledge, our bottom-up domain adaptation approaches outperform the best-performing single-modality pedestrian detection results on KAIST and outperform the state of the art on FLIR.

scholarly journals Shape My Face: Registering 3D Face Scans by Surface-to-Surface Translation

2021 ◽  
Author(s):  
Mehdi Bahri ◽  
Eimear O’ Sullivan ◽  
Shunwang Gong ◽  
Feng Liu ◽  
Xiaoming Liu ◽  
...  
Keyword(s):  
Large Scale ◽  
Training Time ◽  
3D Face ◽  
In The Wild ◽  
Previous State ◽  
Surface Translation ◽  
Visual Attention Mechanism ◽  
Diverse Data ◽  
Human Faces ◽  
Robust To Noise

AbstractStandard registration algorithms need to be independently applied to each surface to register, following careful pre-processing and hand-tuning. Recently, learning-based approaches have emerged that reduce the registration of new scans to running inference with a previously-trained model. The potential benefits are multifold: inference is typically orders of magnitude faster than solving a new instance of a difficult optimization problem, deep learning models can be made robust to noise and corruption, and the trained model may be re-used for other tasks, e.g. through transfer learning. In this paper, we cast the registration task as a surface-to-surface translation problem, and design a model to reliably capture the latent geometric information directly from raw 3D face scans. We introduce Shape-My-Face (SMF), a powerful encoder-decoder architecture based on an improved point cloud encoder, a novel visual attention mechanism, graph convolutional decoders with skip connections, and a specialized mouth model that we smoothly integrate with the mesh convolutions. Compared to the previous state-of-the-art learning algorithms for non-rigid registration of face scans, SMF only requires the raw data to be rigidly aligned (with scaling) with a pre-defined face template. Additionally, our model provides topologically-sound meshes with minimal supervision, offers faster training time, has orders of magnitude fewer trainable parameters, is more robust to noise, and can generalize to previously unseen datasets. We extensively evaluate the quality of our registrations on diverse data. We demonstrate the robustness and generalizability of our model with in-the-wild face scans across different modalities, sensor types, and resolutions. Finally, we show that, by learning to register scans, SMF produces a hybrid linear and non-linear morphable model. Manipulation of the latent space of SMF allows for shape generation, and morphing applications such as expression transfer in-the-wild. We train SMF on a dataset of human faces comprising 9 large-scale databases on commodity hardware.

scholarly journals Enabling rapid COVID-19 small molecule drug design through scalable deep learning of generative models

2021 ◽  
pp. 109434202110109
Author(s):  
Sam Ade Jacobs ◽  
Tim Moon ◽  
Kevin McLoughlin ◽  
Derek Jones ◽  
David Hysom ◽  
...  
Keyword(s):  
Small Molecule ◽  
Molecular Design ◽  
Turnaround Time ◽  
Generative Models ◽  
Quality Model ◽  
Training Time ◽  
Small Molecule Drug ◽  
Training Approach ◽  
Previous State ◽  
Multi Level

We improved the quality and reduced the time to produce machine learned models for use in small molecule antiviral design. Our globally asynchronous multi-level parallel training approach strong scales to all of Sierra with up to 97.7% efficiency. We trained a novel, character-based Wasserstein autoencoder that produces a higher quality model trained on 1.613 billion compounds in 23 minutes while the previous state of the art takes a day on 1 million compounds. Reducing training time from a day to minutes shifts the model creation bottleneck from computer job turnaround time to human innovation time. Our implementation achieves 318 PFLOPs for 17.1% of half-precision peak. We will incorporate this model into our molecular design loop enabling the generation of more diverse compounds; searching for novel, candidate antiviral drugs improves and reduces the time to synthesize compounds to be tested in the lab.

scholarly journals Domain Adaptation Using a Three-Way Decision Improves the Identification of Autism Patients from Multisite fMRI Data

2021 ◽  
Vol 11 (5) ◽  
pp. 603
Author(s):  
Chunlei Shi ◽  
Xianwei Xin ◽  
Jiacai Zhang
Keyword(s):  
Machine Learning ◽  
Domain Adaptation ◽  
Recognition Accuracy ◽  
State Of The Art ◽  
Autism Spectrum ◽  
Fmri Data ◽  
Target Domain ◽  
Sample Distribution ◽  
Machine Learning Methods ◽  
First Time

Machine learning methods are widely used in autism spectrum disorder (ASD) diagnosis. Due to the lack of labelled ASD data, multisite data are often pooled together to expand the sample size. However, the heterogeneity that exists among different sites leads to the degeneration of machine learning models. Herein, the three-way decision theory was introduced into unsupervised domain adaptation in the first time, and applied to optimize the pseudolabel of the target domain/site from functional magnetic resonance imaging (fMRI) features related to ASD patients. The experimental results using multisite fMRI data show that our method not only narrows the gap of the sample distribution among domains but is also superior to the state-of-the-art domain adaptation methods in ASD recognition. Specifically, the ASD recognition accuracy of the proposed method is improved on all the six tasks, by 70.80%, 75.41%, 69.91%, 72.13%, 71.01% and 68.85%, respectively, compared with the existing methods.

scholarly journals Multi-granularity Deep Local Representations for Irregular Scene Text Recognition

2021 ◽  
Vol 2 (2) ◽  
pp. 1-18
Author(s):  
Hongchao Gao ◽  
Yujia Li ◽  
Jiao Dai ◽  
Xi Wang ◽  
Jizhong Han ◽  
...  
Keyword(s):  
State Of The Art ◽  
Visual Representation ◽  
Text Recognition ◽  
Natural Scene ◽  
Attention Network ◽  
Training Time ◽  
Scene Text ◽  
Benchmark Datasets ◽  
Local Representations ◽  
Scene Text Recognition

Recognizing irregular text from natural scene images is challenging due to the unconstrained appearance of text, such as curvature, orientation, and distortion. Recent recognition networks regard this task as a text sequence labeling problem and most networks capture the sequence only from a single-granularity visual representation, which to some extent limits the performance of recognition. In this article, we propose a hierarchical attention network to capture multi-granularity deep local representations for recognizing irregular scene text. It consists of several hierarchical attention blocks, and each block contains a Local Visual Representation Module (LVRM) and a Decoder Module (DM). Based on the hierarchical attention network, we propose a scene text recognition network. The extensive experiments show that our proposed network achieves the state-of-the-art performance on several benchmark datasets including IIIT-5K, SVT, CUTE, SVT-Perspective, and ICDAR datasets under shorter training time.

Sign in / Sign up

Export Citation Format

Share Document