Open Data Science Conference
  • BOOTCAMP
  • EAST
  • WEST
  • EUROPE
  • APAC
  • Ai+
  • Blog
  • Partners
  • Newsletter
  • Jobs
  • About
  • Home
Open Data Science ConferenceOpen Data Science ConferenceOpen Data Science ConferenceOpen Data Science Conference
  • Focus Areas
    • Hands-on Training
    • Deep Learning & Machine Learning
    • Machine Learning for Programmers
    • Data Visualization
    • Data Science Kick Start
    • AI X Business
    • MLOps & Data Engineering
    • Research Frontiers
    • R for Data Science
    • NLP
    • Mini-Bootcamp
  • Bootcamp
    • Register
    • Program Overview
    • Specialization Tracks
    • Bring Your Team
  • AIx Biz
  • Attend
    • Why Attend
    • Convince Your Boss
    • Bring Your Team
    • Download Attendee Guide
    • See Who Attends East
  • Schedule
    • Schedule Overview
    • Preliminary Schedule
    • Speakers
    • Training
  • Register
    • Conference Tickets
    • Bootcamp Tickets
    • AI Expo Tickets
    • AIx Summit
    • Career Expo Tickets
    • Bring Your Team
  • Speakers
    • Call for Speakers
    • East Speakers
  • Partner
    • Partner with ODSC
    • Meet our Partners
    • Partner Brochure
    • Hiring
    • AI Expo Hall
  • Info
    • Media Pass
    • Discounts
    • Volunteer
    • Scholarship Passes
    • Conference Guide
    • FAQ
  • Focus Areas
    • Hands-on Training
    • Deep Learning & Machine Learning
    • Machine Learning for Programmers
    • Data Visualization
    • Data Science Kick Start
    • AI X Business
    • MLOps & Data Engineering
    • Research Frontiers
    • R for Data Science
    • NLP
    • Mini-Bootcamp
  • Bootcamp
    • Register
    • Program Overview
    • Specialization Tracks
    • Bring Your Team
  • AIx Biz
  • Attend
    • Why Attend
    • Convince Your Boss
    • Bring Your Team
    • Download Attendee Guide
    • See Who Attends East
  • Schedule
    • Schedule Overview
    • Preliminary Schedule
    • Speakers
    • Training
  • Register
    • Conference Tickets
    • Bootcamp Tickets
    • AI Expo Tickets
    • AIx Summit
    • Career Expo Tickets
    • Bring Your Team
  • Speakers
    • Call for Speakers
    • East Speakers
  • Partner
    • Partner with ODSC
    • Meet our Partners
    • Partner Brochure
    • Hiring
    • AI Expo Hall
  • Info
    • Media Pass
    • Discounts
    • Volunteer
    • Scholarship Passes
    • Conference Guide
    • FAQ
Oct
09

Deep Learning-Driven Text Summarization & Explainability with Reuters News Data

  • Posted By : odscadmin/
  • 0 comments /
  • Under : Deep Learning, NLP

Image credit: REUTERS/Dominic Ebenbichler for Reuters news data

Editor’s note: At ODSC West 2020, Nadja Herger, Nina Hristozova, and Viktoriia Samatova will hold a workshop focused on text summarization and that will allow you to automatically generate news headlines powered by Reuters News, and learn about the power of transfer learning and explainable AI.

Natural Language Processing (NLP) is one of the fastest-moving fields within AI and it encompasses a wide range of tasks, such as text classification, question-answering, translation, topic modeling, sentiment analysis, and summarization. Here, we focus on text summarization, which is a powerful and challenging application of NLP.

Summarization & Transfer Learning

When discussing summarization, an important distinction to make is between extractive and abstractive summarization. Extractive summarization refers to the process of extracting words and phrases from the text itself to create a summary. Abstractive summarization more closely resembles the way humans write summaries [link]. The key information of the original text is maintained using semantically consistent words and phrases. Due to its complexity, it relies on advances in Deep Learning to be successful [source].

Here, we investigate the automatic generation of headlines from English news articles across all content categories based on the Reuters News Archive, which is professionally produced by journalists and strictly follows rules of integrity, independence and freedom from bias [source]. The headlines themselves are considered fairly abstractive, with over 70% of bi-grams, and over 90% of 3-grams being novel.

We see a trend towards pre-training Deep Learning models on a large text corpus and fine-tuning them for a specific downstream task (also known as transfer learning) [source]. This has the advantage of reduced training time, as well as needing less training data to achieve satisfactory results. Due to the democratization of AI, we observe a leveling of the playing field where everyone can get hold of these models and adapt them for their use cases. We finetuned a state-of-the-art summarization model on Reuters news data, which significantly outperformed the base model itself. An example of a tokenized, unformatted article text and associated machine-generated headline is shown below. The original article text was published by Reuters in October 2019 [link].

Explainability

Do you trust this automatically generated news headline? Researchers commonly rely on the ROUGE score to evaluate the model’s performance for such a task [source]. In its most basic form, it essentially measures the overlap of n-grams between the machine-generated and human-written summaries. If I told you that the model has a ROUGE score of around 45 on the hold-out set, is that sufficient for you to trust the prediction on a previously unseen article text?

How can we increase trust in what the model generated? The move towards more complex models for NLP tasks makes the need for explainable AI more apparent. Explainable AI is an umbrella term for a range of techniques, algorithms, and methods, which accompany outputs from AI systems with explanations [source]. As such, it addresses the often undesired black-box nature of many AI systems, and subsequently allows users to understand, trust, and manage AI solutions. The desired level of explainability depends on the end user [source]. Here, we are interested in making the model output explainable to a potential reviewer rather than for example an AI system builder, who would have different expectations in terms of technical details.

Let us take a look at how adding an explainability feature can support us in our task of verifying if the machine-generated headline is factually accurate. In addition to just generating the headline, we can gain insights into the most relevant parts of the news article. The illustration below builds upon the example shared earlier, by adding highlights to the article text in Reuters news data.

Reuters news data

The darker the highlights, the more important a given word for the resulting headline text. This makes it significantly easier to verify the headline itself. Particularly the first sentence seems to have the largest impact on the generated headline. Interestingly, it refers to “until early next year” rather than “until early 2019”. The year “2019” never occurs in the article text. For this particular example, we actually have access to the human-written headline, see screenshot of the original article from the Reuters site below.

Reuters news dataKnowing that the article was published in 2019, it is evident that “early next year” refers to the year 2020 rather than 2019, and thus renders the machine-generated headline partially inaccurate. We believe that verifying machine-generated headlines with an extra layer of explainability leads to increased trust and easier detection of biases or mistakes.

For more details on text summarization and Reuters news data, the power of transfer learning, as well as adding explainability for increased trust, join us for our hands-on workshop at ODSC West in October. You will walk away with an interactive notebook to get a head start in applying these concepts to your own challenges!


Nadja Herger is a Data Scientist at Thomson Reuters Labs, based in Switzerland. She is primarily focusing on Deep Learning PoCs within the Labs, where she is working on applied NLP projects in the legal and news domains, applying her skills to text classification, metadata extraction, and summarization tasks.

 

 

 

Nina Hristozova is a Data Scientist at Thomson Reuters (TR) Labs. She has a BSc in Computer Science from the University of Glasgow, Scotland. As part of her role at TR she has worked on a wide range of projects applying ML and DL to a variety of NLP problems. Her current focus is on applied summarization of legal text.

 

 

 

Viktoriia Samatova is a Head of Applied Innovation team of Data Scientists within Reuters Technology division focused on discovering and applying new technologies to enhance Reuters products and improving efficiency of news content production and discoverability.


Oct
02

Why I Love Keras and Why You Should Too

  • Posted By : odscadmin/
  • 0 comments /
  • Under : Deep Learning

I started working with Deep Learning (DL) in the 2016 – 2017 time frame when the framework ecosystem was much more diverse and fragmented than it is today. Theano was the gold standard at the time, Tensorflow had just been released, and DeepLearning4j was still being built. Both Theano and Tensorflow were relatively low-level frameworks, and were somewhat painful to work with, especially for newbies like me. The Keras and Lasagne libraries helped alleviate the pain somewhat, offering a higher level API — Lasagne wrapped Theano, while Keras provided a common interface for either Theano or Tensorflow.

[Related article: Deep Learning in R with Keras]

My first DL project was Image Classification. By this time, the ImageNet problem was pretty much considered solved, with submitted ML models routinely scoring in excess of 95% accuracy. For reference, the ImageNet task was to distinguish between about 1000 types of everyday objects. My classifier, on the other hand, needed to distinguish between 8 (later 11) classes of medical images.

Using Transfer Learning to leverage the knowledge in the ImageNet models seemed like a natural fit. In my case, Transfer Learning would involve taking an existing high-performing ImageNet model, and fine-tuning it with my labeled data to fit the medical image classification task.

The first iteration of my model ended up being a hybrid of a Caffe backend to generate vector image representations using trained ImageNet models, and a Keras Dense network to classify these vectors into one of 8 classes. Caffe is a C++ framework specialized for image tasks, and the main reason I chose it was because it provided downloadable trained ImageNet models. Later, when the Keras project provided its own downloadable ImageNet models, I built a second iteration of the model, which fine-tuned the ImageNet model and learned to classify medical images end-to-end.

I bring up this story to illustrate how easy and natural it felt to start working with Keras. For a long time, that was the only library I needed for building my models. Over this period, I have been consistently delighted by its intuitive API, its sensible default parameters, and the overall quality of tutorial material on its website.

However, if you are like me, you often have to understand and train other peoples’ models in addition to your own. Since these models are often built by DL researchers using Pytorch and Tensorflow, I ended up learning these frameworks as well. I can attest from experience that Keras is by far the easiest to learn and use.

Today, the Deep Learning ecosystem is much more mature, so thankfully one can get by with learning fewer frameworks. While many excellent frameworks have been released over these intervening years, and are being used in specialized niches, the major ones are Keras, Tensorflow, and Pytorch. Pytorch became popular because of its eager execution model, which Tensorflow did not allow, and which Keras hid behind its cleverly-designed API. Keras has since been subsumed into Tensorflow as tf.keras, but the original Keras lives on as well, with an additional CNTK (from Microsoft) backend. For its part, Tensorflow, in its 2.x incarnation, has embraced Pytorch’s eager execution model, and made tf.keras its default API.

So there has been lots of convergence, and while I recommend learning all three of the frameworks listed above, if you need to get productive quickly and don’t have a framework imposed upon you (by the project or by corporate policy), I would recommend you start with Keras. If you are proficient in one of the others, you should still learn Keras, because for a majority of tasks you are likely to be more productive with Keras than with your current framework.

Of course, the simplicity and elegance of Keras comes at a price. Most things are easy and intuitive to do in Keras, but certain things are very hard or even impossible. Some of these tasks are possible in the other lower-level frameworks, but you pay for that convenience with more verbose code and a steeper learning curve. However, with tf.keras (and to some extent with the original Keras using Tensorflow backend), you have access to the full Tensorflow substrate. In addition, the Keras team has recently been busy refining their code, which now makes it possible to do certain things in Keras that were previously thought to be impossible.

I am honored to present Keras: from soup to nuts – an example-driven tutorial at OSDC West this year, where I hope to touch upon some of these things that make Keras such a simple yet powerful addition to your DL toolbox. I hope to see you there!


About the author/ODSC West speaker: Sujit Pal is an applied data scientist at Elsevier Labs, an advanced technology group within the Reed-Elsevier Group of companies. His areas of interests include Semantic Search, Natural Language Processing, Machine Learning, and Deep Learning. At Elsevier, he has worked on several machine learning initiatives involving large image and text corpora, and other initiatives around recommendation systems and knowledge graph development. He has co-authored Deep Learning with Keras (https://www.packtpub.com/big-data-and-business-intelligence/deep-learning-keras) and Deep Learning with Tensorflow 2.x and Keras (https://www.packtpub.com/data/deep-learning-with-tensorflow-2-0-and-keras-second-edition), and writes about technology on his blog Salmon Run (https://sujitpal.blogspot.com/).


Jul
07

Image Detection as a Service

  • Posted By : odscadmin/
  • 0 comments /
  • Under : Deep Learning

Across our two brands, Badoo and Bumble, we have over 500 million registered users worldwide uploading millions of photos a day to our platform. These images provide us with a rich data set from which we derive a wealth of insights.

User Profile Information

Within our data science team, we created a service where images are input, and output is the image content information. We use this service during our prototyping phase, during our exploratory analysis and for delivering ad-hoc insights to the business about image content.

For example, we have observed that if someone is wearing sunglasses in all of their pictures, they tend to receive fewer likes than users who clearly show their faces. This enables us to provide tips to users on how to enhance their profiles.

In this blog, I give an overview of how we combined deep neural networks and Flask APIs to offer this service.

Computer Vision Tasks

Our API serves a variety of models, providing different types of information on image content. Some of the models we use are:

  • Perform image classification
  • Provide textual descriptions of image content
  • Object detection

Impact & Workflow

During the execution of projects, we carry out a number of potential impact analyses. These give an estimate of the value we expect the project to deliver as we learn more and progress throughout the process. The purpose of this is to ensure that resources get allocated effectively and that what we produced delivers maximum impact on the business. The visual below outlines the workflow we aim for.

https://odsc.com/europe/Project Workflow

During the prototyping phase of our computer vision models, we need to be able to assess what the possible impact might be on the business if we were to allocate resources and productionise them. In order to assess this fully, we needed to be able to scale the number of images we could score. To this end, we built a Flask framework to enable us to serve the models on a greater scale compared to using a local machine.

Web API

Once we had trained our models using Jupyter notebooks and .py scripts we wanted other members of the team and people across the business to be able to use them to support their prototyping efforts and potential impact reviews. To achieve this, we decided to encapsulate the models in REST APIs. An API essentially allows you to interact over HTTP, making requests to specific URLs and getting relevant data back in the response.

Why APIs?

The reason we decided to use APIs is that it makes it easy for cross-language applications to work well. For example, when it’s necessary for a front-end developer to use these models, they simply need to get the endpoint of the API and have no need to be familiar with Python or have domain-specific knowledge.

There are a host of third-party solutions offering machine vision APIs including Google Cloud Vision and AWS Rekognition. We decided against going down this route in the interests both of minimising costs and keeping our data in-house. We used Python Flask to build and serve our API in-house. Flask is a microframework for Python and offers a powerful way of annotating Python functions with REST endpoints.

Why Flask?

Flask and Django are relatively comparable to Python web frameworks. We decided to use Flask over Django because it is very simple and easy to get started with whereas Django is quite heavy for building web applications. Simplicity and flexibility being two key requirements for our service also influenced our decision.

Hosting the Service

Once our Flask API was up and running on a local machine, we then packaged up the service as an application on one of our servers for easy access by other people in the business. On these servers, we have GPUs which help to accelerate the computational time.

Docker

In order to containerize the service on our server, we created a Docker container from an image. If you are not familiar with Docker I would recommend taking a look at their documentation online as it is very thorough and digestible.

Summary

The infrastructure we have developed is primarily used by the Data Science team when carrying out exploratory pieces of work, performing ad-hoc analysis, and during our model prototyping phase. We also use the service to help to determine whether we should be investing resources into putting the models onto production. We have found that the framework allows our team to work in a more unified yet flexible way.

During my talk at ODSC, “Image Detection as a Service: How we Use APIs and Deep Learning to Support our Products,”  I’ll discuss this service in more detail, the challenges we faced along with some advice regarding best practices. I hope you will attend and enjoy the talk.

[Enhancing Discovery in Data Science Through Novelty in Machine Learning]


About the author/ODSC Speaker: Laura Mitchell

With over 10 years of experience in the tech and data science space, I am the Lead Data Scientist at MagicLab whose brands, Badoo and Bumble, have connected the lives of over 500 million people through dating, social and business.

I am a published author in the field of Artificial Intelligence and have hands-on and leadership experience in the delivery of projects using natural language processing, computer vision, and recommender systems, from initial conception through to production.


Clinical Development
Feb
24

East 2020 Preview: Impacting Clinical Development with Advanced Analytics: Challenges & Opportunities

  • Posted By : odscadmin/
  • 0 comments /
  • Under : Deep Learning

Jennifer is a speaker for ODSC East 2020 this April 13-17 in Boston. Be sure to check out her talk, “Impacting Clinical Development with Advanced Analytics: Challenges & Opportunities,” there!

Why did I become a statistician in Big Pharma? When I completed my training, I had never considered it. Big Pharma statisticians were generally considered pencil pushers, engaged in the most boring work possible. But when a position opened up as a research statistician at Genentech (which is a member of the Roche group), I wanted it – because Genentech has always had the reputation of being at the forefront of biological science and research, in particular, seemed more promising in terms of interesting work. With Roche’s acquisitions of Flatiron Health and Foundation Medicine and partnerships like PicnicHealth, science is now more broadly defined to include data science as well. With this large-scale transformation occurring within the company, opportunities to pursue methodology development on novel data types now exist in all functions of the company – embedded statisticians are in turn uniquely positioned to directly impact our pipeline.

Pharma generally recognizes the need to embrace change in how we analyze and handle data internally, a process in the past clinical teams did not prioritize as the end result of a trial is meant to be a filing, not a curated data mart. Now that we are focused on curated and integrated datasets moving forward, there is naturally a commensurate pressure to gain additional insights from this data. Expectations are higher – we’ve already run the logistic regression or mixed effect model repeat measurement (MMRM) on the primary, secondary and exploratory endpoints. In order to justify the heavy-lift needed to wrangle curated data from our incurred technical debt, the algorithm used to analyze this data must be proportionally complex. Terms start flying around like advanced analytics, machine learning, deep learning, and artificial intelligence.

What is the bar to adopt these analytical methodologies?  The core of Big Pharma is to develop drugs and transition new molecules through the pipeline of progressing evidence in support of helping patients. We expect that our clinical development plans are informed by data and evidence. Just like any modeling exercise – there are training and test data. In this setting, training data is a large non-interventional cohort, which does not typically reflect the highly ascertained patient population of a trial. As an example, this atezolizumab trial in Non-Small Cell Lung Cancer has 35 inclusion/exclusion criteria, many of which are based on variables that will not typically be collected in an observational cohort or electronic health medical record database. This is simply the first obstacle – but assuming that we will hit the bar, what are the rest of the obstacles unique to Pharma that we will need to address? What do our first forays at the forefront of integrating data science into our pipelines look like? Come find out at my talk where I would love to hear your feedback!


More on the author/speaker:

Jennifer Tom joined Genentech in 2014 as a statistician in bioinformatics and computational biology and moved into product development in 2018. She has supported human genetics, microbiome, imaging, early clinical development, and biomarker activities across various non-oncology indications. Previously she worked as a software engineer at Agilent and the visiting assistant Neyman Professor of statistics at Berkeley. Jennifer received a BA in Molecular and Cellular Biology from UC Berkeley and an MS and PhD in Biostatistics from UCLA.


Feb
12

A Brief Survey of Node Classification with Graph Neural Networks

  • Posted By : odscadmin/
  • 0 comments /
  • Under : Deep Learning

Shauna is a speaker for ODSC East 2020 this April 13-17 in Boston. Be sure to check out her talk, “Graph Neural Networks and their Applications,” there!

Graph neural networks have revolutionized the performance of neural networks on graph data. Companies such as Pinterest[1], Google[2], and Uber[3] have implemented graph neural network algorithms to dramatically improve the performance of large-scale data-driven tasks.

Introduction to Graphs

A graph is any dataset that contains nodes and edges. Nodes are entities (e.g. people, organizations), and edges represent the connections between nodes. For example, a social network is a graph in which people in the network are considered nodes. Edges exist when two people are connected in some way (e.g. friends, sharing one’s posts). 

In retail applications, customers and products can be viewed as nodes. An edge shows the relationship between the customer and a purchased product. A graph can be used to represent spending habits for each customer. Additionally, nodes can also have features or attributes. People have attributes such as age and height, and products have attributes such as price and size. Pinterest has used graph neural networks in this fashion to improve the performance of its recommendation system by 150%[1].

The advent of Graph Neural Networks

Until the development of graph neural networks, deep learning methods could not be applied to edges to extract knowledge and make predictions, and instead could only operate based on the node features. Applying deep learning to graph data allows us to approach tasks link prediction, community detection, and generating recommendations. 

Deep learning can also be applied to node classification, or predicting the label of an unlabelled node. This takes place in a semi-supervised setting, where the labels of some nodes are known, but others are unknown. A survey of deep learning node classification methods shows a history of advances in state-of-the-art performance while illustrating the range of use cases and applications. 

Methods discussed in this blog post were evaluated on the benchmark CoRA dataset. CoRA consists of journal publications in deep learning. Each publication is a node, and edges exist between nodes if they cite or are cited by another paper in the dataset. The dataset is comprised of 2708 publications with 5429 edges. 

DeepWalk

DeepWalk[4], released in 2014, was the first significant deep learning-based method to approach node classification. DeepWalk’s approach was similar to that taken in natural language processing (NLP) to derive embeddings. An embedding is a vector-representation of an object such a word in NLP or a node in a graph. To create its embeddings, DeepWalk takes truncated random walks from graph data to learn latent representations of nodes. On the CoRA dataset, DeepWalk achieved 67.2% accuracy on the benchmark node classification experiment[5]. At the time, this was 10% better than competing methods with 60% less training data.

To demonstrate how the embeddings generated by DeepWalk contain information about the graph structure, below is a figure that shows how DeepWalk works on Zachary’s Karate Network. The Karate Network is a small network of connections of members of a karate club, where edges are made between two members if they interact outside of karate. The node coloring is based on the different sub-communities within the club. The left image is the input graph of the social network, and the right is the two-dimensional output generated by the DeepWalk algorithm operating on the graph data. 

Source: [4].

Graph Convolutional Networks

In 2016, Thomas N. Kipf and Max Welling introduced graph convolutional networks (GCNs)[6], which improved the state-of-the-art CoRA benchmark to 81.5%. A GCN is a network that consists of stacked linear layers with an activation function. Kipf and Welling introduced a new propagation function that operates layer-wise and works directly on graph data. 

Source: [6] The t-SNE visualization of the two-layer GCN trained on the CoRA dataset using 5% of labels. The colors represent document class. 

The number of linear layers in a GCN determines the size of the target node neighborhood to consider when making the classification prediction. For example, one hidden layer would imply that the graph network only examines immediate neighbors when making a classification decision. 

The input to a graph convolutional network is an adjacency matrix, which is the representation of the graph itself. It also takes the feature vectors of each node as input. This can be as simple as a one-hot encoding of each node’s attributes, while more complex versions can be used to represent the complex features of a node. 

Applying GCN to Real-World Data

Our research team was interested in implementing a GCN on real-world data in order to speed up analyst tasking. We implemented a GCN architecture written in PyTorch to perform node classification on article data. The graph used in our dataset was derived from article data grouped together in “playlists” by a user who determined that they were relevant. The nodes were individual articles and playlists, and edges existed between an article and a playlist if a specific article was included in the playlist. Instead of manually poring through the corpus to determine additional relevant articles, we used the GCN to recommend other potentially relevant documents. After running our corpus of about 100,000 articles and 7 different playlists through a 2-layer GCN, our network performed 5x better than random.

Graph-BERT

GCN’s were the leading architecture for years, and many variations of them were subsequently released. Then, in January 2020 Graph-BERT[7] removed the dependency on links and reformatted the way that graph networks are usually represented. This is important for scalability, while also showing improved accuracy and efficiency relative to other types of graph neural networks. We are currently exploring how Graph-BERT will impact the use cases we have been addressing with graph neural networks.

Conclusion 

Graph neural networks are an evolving field in the study of neural networks. Their ability to use graph data has made difficult problems such as node classification more tractable. 

For a deeper discussion on graph neural networks and the problems that they can help solve, attend my talk at ODSC East, “Graph Neural Networks and their Applications.”

CITATIONS

[1] R. Ying, R. He, K. Chen, P. Eksombatchai, W. L. Hamilton, and J. Leskovec, “Graph convolutional neural networks for web-scale recommender systems,” in Proc. of KDD. ACM, 2018, pp. 974–983.

[2]  Sanchez-Lengeling , Benjamin, et al. “Machine Learning for Scent: Learning Generalizable Perceptual Representations of Small Molecules.” ArXiv, no. 1910.10685, 2019. Stat.ML, arxiv.org/abs/1910.10685.

[3] Uber Engineering. “Food Discovery with Uber Eats: Using Graph Learning to Power Recommendations”, eng.uber.com/uber-eats-graph-learning.

[4] Perozzi, Bryan, Al-Rfou, Rami and Skiena, Steven. “DeepWalk: Online Learning of Social Representations.” CoRR abs/1403.6652 (2014).

[5] “Node Classification on CoRA.” paperswithcode.com/sota/node-classification-on-coraom.

[6] Kipf, Thomas N. and Welling, Max. “Semi-Supervised Classification with Graph Convolutional Networks.” Paper presented at the meeting of the Proceedings of the 5th International Conference on Learning Representations, 2017.

[7] Zhang, Jiawei, Zhang, Haopeng, Xia, Congying, and Sun, Li. “Graph-Bert: Only Attention is Needed for Learning Graph Representations.” arxiv.org/abs/2001.05140 (2020).


About the speaker/author: Shauna Revay is an applied machine learning researcher for Novetta’s Machine Learning Center of Excellence. She specializes in rapid prototyping of ML solutions spanning fields such as NLP, audio analysis and ASR, and the implementation of graph neural networks. Shauna earned her PhD in mathematics at George Mason University where she studied harmonic and Fourier analysis. She pursues research in interdisciplinary topics and has published papers in mathematics, biology, and electrical engineering journals and conference proceedings.


123
Categories
  • Accelerate AI (5)
  • Career (4)
  • Data Science (41)
  • Data Visualization (5)
  • Deep Learning (12)
  • Machine Learning (37)
  • NLP (10)
  • Python (4)
  • R (1)
  • Statistics (1)
Recent Posts
  • Top ODSC Europe 2020 Sessions Available for Free On-Demand October 9,2020
  • Announcing ODSC APAC Dec 8-9 October 9,2020
  • Announcing the ODSC Ai x West Business and Innovation Summit This Oct 29-30 October 9,2020
Open Data Science

Open Data Science
Innovation Center
101 Main St
Cambridge, MA 02142
info@odsc.com

Menu
  • Partner with ODSC
  • Blog
  • Training
  • Jobs
  • FAQ
Conferences
  • East 2021
  • West 2021
  • Europe 2021
  • APAC 2020
Extras
  • Newsletter
  • About
  • Code of Conduct
  • Privacy Policy
Copyright ODSC 2020. All Rights Reserved
Close