• ChatBots – the next generation of messaging apps

    ******* Enter the Copley Place Mall (right next to the Marriott) and follow signs to the "Offices." The correct escalator is next to Barneys New York, or take the elevator to the Sky Lobby level. Please bring photo identification and check in at the desk in the lobby, to pick up your name tag. Then, take an elevator to the 7th floor.******* Agenda: 6:30 - Doors open. Networking. Beer & snacks. 6:45 - Opening remarks. 7:00 - Chatbots with NLP by Slater Victoroff 7:20 - Q&A break 7:30 - Optimizing Business Strategies With NLP and Machine Learning by Michael Frederikse 7:50 - Q&A break 8:00 - Wrap-up _________________________________________________________ DETAILED AGENDA: Speaker: Slater Victoroff Title: Chatbots with NLP Abstract: Chatbots are one of the most highly-hyped new consumer technologies being enabled by modern deep learning. However the hype and the reality often do not connect. Many solutions that are pitched as "chatbots" are actually run by human beings behind an API, and many "chatbot" systems are no better than keyword-based routing. Come learn about what is and is not possible when it comes to chatbots. Learn how to approach chatbot systems from a practical perspective and learn enough machine learning to understand how to break down and approach a chatbot system. Bio: Slater Victoroff is the Founder and CTO of indico data solutions, an Enterprise AI solution for unstructured content with an emphasis on text and NLP. He has been building machine learning solutions for startups, governments, and fortune 100 companies for the past 5 years. _______________________________________________________ Speaker: Michael Frederikse Title: Optimizing Business Strategies With NLP and Machine Learning Abstract: The goal of this presentation is to layout strategies and methods to accelerate your business processes. Starting from a friendly introduction into Natural Language Processing (NLP) and the many opaque buzzwords surrounding it, we’ll dive into some of the technical methods and approaches. Equipped with these tools, we will cover several case studies that will help showcase possible implementations to optimize and expand your business and/or reporting practices. Lastly, we will open up for questions about specific solutions that you might be looking to introduce in your own setting. Bio: Michael Frederikse is a Software Technologist at Pattern Inc, a boutique consulting firm specializing in machine learning and artificial intelligence solutions. Through consulting, he's had the pleasure to work in a number of different industries including defense/intelligence, political science, international shipping/supply chain, pharmaceuticals/life sciences, and investment finance. Most recently, he has been engaged with Loomis Sayles as a Principal ML Engineer, helping to develop novel AI and ML approaches to complement their investment strategies. Please register here https://bit.ly/2qo7ReG

  • Boston meetup: Deep Dive into TensorFlow #5

    Amazon Cambridge Offices

    ***WAITLIST ONLY*** PLEASE REGISTER HERE: http://bit.ly/2qZn9aq Many thanks to Alexa, an Amazon Company (http://www.alexa.com/) for sponsoring the TensorFlow meetup! Agenda: 6:30 - Doors open. Networking. Beer and Pizza 7:00 - Introduction to Probabilistic Modeling with the TensorFlow-based Stochastic Variational Bayesian Inference Library Edward by Alex Coventry 7:40 - Q&A break 7:50 - TensorFlow to Predict Housing Prices by Sam Putnam 8:30 - Q&A break 8:40 - Wrap-up DETAILED AGENDA: Introduction to Probabilistic Modeling with the TensorFlow-based Stochastic Variational Bayesian Inference Library Edward Edward (http://edwardlib.org/) is a fast, flexible statistical inference tool based on optimization methods commonly applied to training Deep Neural Nets. In this brief introduction we'll discuss how it works and how to use it. Speaker: Alex Coventry, Mathematician and Machine Learning Engineer Bio: Alex Coventry (@AlxCoventry) is a Mathematician and Machine Learning Engineer with extensive experience in Bayesian Inference and Deep Learning Methods. With his friend Paul Miller he runs the Cambridge AI Meetup (https://www.meetup.com/Cambridge-Artificial-Intelligence-Meetup/), which meets weekly as the SIPB Deep Learning Reading Group to discuss Deep Learning papers of interest to the group. (See a list of papers we've discussed here (https://github.com/pmiller10/cambridge-ai).) He is developing a method which libraries like Edward will be able to use for more complete estimates of statistical uncertainty. ******************************************************** TensorFlow to Predict Housing Prices Speaker: Sam Putnam, CEO/Founder and Deep Learning Consultant at Enterprise Deep Learning. Bio: Sam Putnam is CEO/Founder, and a Deep Learning Consultant, at Enterprise Deep Learning. He directs machine learning projects, solves clients' business problems, and trains deep neural networks on large datasets. Sam is also a contributor to the TensorFlow machine learning project and a member of the Machine Learning Society. Sam previously worked at the Dartmouth Center for Imaging Medicine, the National Renewable Energy Laboratory, and the Lab for Cognition and Control in Complex Systems.

  • ONLINE WEBINAR: Text Prediction Using Recurrent Neural Networks with TensorFlow

    *Please register HERE (https://goo.gl/l9E7vq)to get the unique link to join the webinar* Are you willing to learn how to perform text prediction using deep learning and TensorFlow? Join the webinar (https://goo.gl/l9E7vq) to learn more! Overview In this webinar, Dipendra Jha (https://www.linkedin.com/in/dipendra009/?__hstc=38408444.8eca82a5236137fe83a75da41d85c24f.1488281537209.1491916684542.1491992597887.40&__hssc=38408444.5.1491992597887&__hsfp=3830342050) and Reda Al-Bahrani (https://www.linkedin.com/in/redaalbahrani?__hstc=38408444.8eca82a5236137fe83a75da41d85c24f.1488281537209.1491916684542.1491992597887.40&__hssc=38408444.5.1491992597887&__hsfp=3830342050) will demonstrate how to use Sequence-to-sequence decoder from TensorFlow library build using Long Short Term Memory (LSTM) cells, on top of the text inputs embedded using the embedding lookup available in TensorFlow. You will learn about: - the required steps to accomplish text prediction from input processing, embeddings, to using LSTMs to make the prediction - the concepts of Recurrent Neural Networks (RNNs) and Long Short Term Memory (LSTMs), followed by coding and demonstration for accomplishing the acutal text prediction using TensorFlow. Who should attend? This webinar will be of interest to Data Scientists, Software Engineers and Entrepreneurs in the areas of Connected Cars, Internet of Things/Industrial Internet, Medical Devices, Financial Technology (blockchain) and predictive apps/APIs of all sorts. About the Presenters: Dipendra Jha (https://www.linkedin.com/in/dipendra009/?__hstc=38408444.8eca82a5236137fe83a75da41d85c24f.1488281537209.1491916684542.1491992597887.40&__hssc=38408444.5.1491992597887&__hsfp=3830342050) is a fourth-year Ph.D. Candidate in Computer Engineering at Northwestern University. He is exploring the field of Deep Learning and Machine Learning using High Performance Computing (HPC) systems in the CUCIS lab under Prof. Alok Choudhary. His research focuses on scaling up deep learning and machine learning models using HPC system, and their application to accelerate Materials Discovery in the field of Materials Science and Engineering. Prior to this, he completed his Master’s in Computer Science from Northwestern University. Reda Al-Bahrani (https://www.linkedin.com/in/redaalbahrani/?__hstc=38408444.8eca82a5236137fe83a75da41d85c24f.1488281537209.1491916684542.1491992597887.40&__hssc=38408444.5.1491992597887&__hsfp=3830342050) is a Ph.D. Candidate in Computer Science from Northwestern University. He is exploring the field of Deep Learning and Machine Learning in the CUCIS lab under Prof. Alok Choudhary. His research focuses on knowledge discovery for health informatics from structured data and textual data. He worked in IT infrastructure support at Saudi Aramco and on XHQ an Operations Intelligence Software at Siemens Saudi Arabia before joining the CUCIS lab. Before this, he completed his Master’s in E-Commerce Technology from DePaul University.

  • Deep Dive into TensorFlow #4

    DataXu Inc

    ***WAITLIST ONLY*** Please register using EVENTBRITE.COM (http://bit.ly/2oC3utl) Many thanks to DataXu (https://www.dataxu.com/)for hosting and sponsoring the TensorFlow meetup! Agenda: 6:30 - Doors open. Networking. Beer and Pizza 7:00 - Deep Reinforcement Learning with TensorFlow by Lex Fridman 8:00 - Q&A break. 8:15 - Wrap-up. DETAILED AGENDA: Deep Reinforcement Learning with TensorFlow This talk with introduce deep reinforcement learning including its applications, challenges, and recent advances. Code examples in TensorFlow and ConvnetJS will be shown as part of presenting an illustrative deep RL world of DeepTraffic: a trajectory planning competition in a micro-traffic simulation. See http://selfdrivingcars.mit.edu/deeptrafficjs to try before the meeting. Speaker: Lex Fridman, Postdoctoral Associate at MIT Bio: Lex Fridman is a postdoc at MIT, developing and applying new computer vision and deep learning approaches in the context of self-driving cars with a human-in-the-loop. His work focuses on messy, large-scale, real-world data, with the goal of building intelligent systems that have real world impact. Lex received his BS, MS, and PhD from Drexel University where he worked on applications of machine learning, computer vision, and decision fusion techniques in a number of fields including robotics, active authentication, activity recognition, and optimal resource allocation on multi-commodity networks. Before joining MIT, Lex was at Google working on machine learning and decision fusion methods for large-scale behavior-based authentication.

  • ONLINE WEBINAR: Image Classification Done Simply Using Keras and TensorFlow

    *Please register HERE (https://goo.gl/IIaqKP)to get the unique link to join the webinar* Are you willing to learn how to build an image classifier using Keras with a TensorFlow backend? Join the webinar (https://goo.gl/IIaqKP) to learn more! Overview The fact that computers can see is just not that amazing anymore. But, the techniques for teaching a computer to do this are now simpler and more refined than ever. In this webinar, Rajiv Shah (https://www.linkedin.com/in/rcshah?authType=NAME_SEARCH&authToken=TVD5&locale=en_US&srchid=854662381470146485173&srchindex=1&srchtotal=1&trk=vsrp_people_res_name&trkInfo=VSRPsearchId%3A854662381470146485173%2CVSRPtargetId%3A15452467%2CVSRPcmpt%3Aprimary%2CVSRPnm%3Atrue%2CauthType%3ANAME_SEARCH) will describe the process of building an image classifier using Keras with a TensorFlow backend and discuss how to extend the code to your own pictures to make a custom image classifier. The approach here uses Keras, which is emerging as the best library for building neural networks. The code here also assumes you are using TensorFlow as the underlying library. The presentation will give a basic understanding of image classification and show the techniques used in industry to build image classifiers. You will learn: - How to build a simple convolutional network - How to augment the data - How to use a pretrained network - How to use transfer learning by modifying the last few layers of a pretrained network The classification will be based on the classic example of classifying cats and dogs. The code for the presentation can be found here (https://github.com/rajshah4/image_keras). About the Presenter Rajiv Shah (https://www.linkedin.com/in/rcshah/) is a senior data scientist at Caterpillar and an Adjunct Assistant Professor at the University of Illinois at Chicago. Rajiv is an active member of the data science community in Chicago with an interest into public policy issues, such as surveillance in Chicago. He has a PhD from the University of Illinois at Urbana Champaign. You find more of his projects at www.rajivshah.com (https://rajivshah.com/).

    4
  • ONLINE WEBINAR: Deep Learning Using TensorFlow and TensorFlow-Slim

    *****************WAITLIST ONLY******************* *******PLEASE REGISTER HERE (http://paas.ly/2fCxUf8)****** Do you want learn how CNNs work and how to build and train such networks? Join the webinar (http://paas.ly/2fCxUf8) to learn more! In this webinar, Dipendra Jha (https://www.linkedin.com/in/dipendra009?authType=OUT_OF_NETWORK&authToken=k4Fx&locale=en_US&srchid=854662381478188385692&srchindex=1&srchtotal=2413&trk=vsrp_people_res_name&trkInfo=VSRPsearchId%3A854662381478188385692%2CVSRPtargetId%3A194620481%2CVSRPcmpt%3Aprimary%2CVSRPnm%3Atrue%2CauthType%3AOUT_OF_NETWORK), Ph.D. student in Computer Science from Northwestern University, will provide a brief introduction to Deep Learning and TensorFlow, followed by actual implementation and demonstration of MNIST image classification using convolutional neural networks (CNNs). Agenda: Introduction to the Fundamentals of Deep LearningThe Strengths of Using TensorFlowImage Classification Using CNNs for MNIST DatasetHow CNNs work and How to Build and Train Such NetworksThe Usage of TensorFlow for large-scale application of Deep Learning to Big Datasets in IndustryQ&A Join the webinar to (http://www.altoros.com/blog/event/deep-learning-using-tensorflow-and-tensorflow-slim/#get_record): Learn more about the fundamentals of deep learning, followed by the strengths of using TensorFlow Look about image classification using CNNs for MNIST datasetDiscover how CNNs work and how to build and train such networks Examine how TensorFlow can be used for large-scale application of deep learning to big datasets in industry Who should attend: This webinar will be of interest to Data Scientists, Software Engineers and Entrepreneurs in the areas of Connected Cars, Internet of Things/Industrial Internet, Medical Devices, Financial Technology (blockchain) and predictive apps/APIs of all sorts. About the presenter: Dipendra Jha (https://www.linkedin.com/in/dipendra009?authType=OUT_OF_NETWORK&authToken=k4Fx&locale=en_US&srchid=854662381478188385692&srchindex=1&srchtotal=2413&trk=vsrp_people_res_name&trkInfo=VSRPsearchId%3A854662381478188385692%2CVSRPtargetId%3A194620481%2CVSRPcmpt%3Aprimary%2CVSRPnm%3Atrue%2CauthType%3AOUT_OF_NETWORK) is is a fourth-year Ph.D. student in Computer Science from Northwestern University. He is exploring the field of Deep Learning and Machine Learning using High Performance Computing (HPC) systems in the CUCIS lab under Prof. Alok Choudhary. His research focuses on scaling up deep learning and machine learning models using HPC (CPU and GPU) clusters, and their application to Material Science and Social Media Analytics. Before this, he completed his Master’s in Computer Science from Northwestern University. He worked in the field of Computer Networks, Distributed Systems and Cellular Networks in Aqualab under Prof. Fabian Bustamante. During this period, his research spanned from Web Page Performance Optimizations, Network Measurements and Community WiFi to Inter-domain Routing in Cellular Networks, IXPs and Content Distribution Networks (CDNs). He completed his Bachelors’ in Computer Engineering from Tribhuvan University in Nepal.

  • Deep Dive into TensorFlow #3

    Bocoup

    ***WAITLIST ONLY*** Due to security concerns we need the First and Last name, email and company name. Please register using EVENTBRITE.COM (https://www.eventbrite.com/e/boston-meetup-deep-dive-into-tensorflow-3-tickets-28685238320) Many thanks to Bocoup (https://bocoup.com/) for hosting the TensorFlow meetup! Agenda: 6:30 - Doors open. Networking. Members meet each other. Beer and Pizza 7:00 - Building a Self-Driving Car Brain with TensorFlow by Lex Fridman 8:00 - Q&A break. 8:20 - Wrap-up. DETAILED AGENDA: Building a Self-Driving Car Brain with TensorFlow Lex will introduce the main challenges of building a self-driving car and how deep learning (and TensorFlow) can be used to solve each. After that Lex will provide a TensorFlow tutorial for training an end-to-end network that can drive a car using just a video stream from a single camera. Speaker: Lex Fridman, Postdoctoral Associate at MIT Bio: Lex Fridman is a postdoc at MIT, developing and applying new computer vision and deep learning approaches in the context of self-driving cars with a human-in-the-loop. His work focuses on messy, large-scale, real-world data, with the goal of building intelligent systems that have real world impact. Lex received his BS, MS, and PhD from Drexel University where he worked on applications of machine learning, computer vision, and decision fusion techniques in a number of fields including robotics, active authentication, activity recognition, and optimal resource allocation on multi-commodity networks. Before joining MIT, Lex was at Google working on machine learning and decision fusion methods for large-scale behavior-based authentication. ***WAITLIST ONLY***

  • Deep Dive into TensorFlow

    Booz Allen Hamilton

    Please confirm your attendance! Due to security concerns we need the First and Last name - as on your government issued ID registered using EVENTBRITE.COM (https://www.eventbrite.com/e/boston-meetup-tensorflow-use-cases-tickets-25018027594) Many thanks to Booz Allen Hamilton (http://www.boozallen.com/) for hosting this meetup at their wonderful office. We also thank Boston Immersive (http://www.bim.ai)- our food & drinks sponsor – for providing beer and pizza for the meetup. Agenda: 6:30 - Doors open. Networking. Members meet each other. 6:45 - Welcome. Members to share topics of interests for future meetings. 7:00 - Talk #1. An overview on recent models built with Tensorflow by Jason Toy 7:30 - Q&A break. 7:45 - Talk #2. Occupancy Dynamics: A Simple Solution with Tensorflow by Mark Kurtz 8:15 - Q&A break & wrap-up. DETAILED AGENDA : An overview on recent models built with Tensorflow Jason Toy is the founder and CEO of Somatic, a platform for anyone to easily build deep learning applications. He has been building software companies and products from the ground up for 10 years. Jason considers himself a generalist who works wherever the core problems are: programming, sales, training, management, marketing, and DevOps. Occupancy Dynamics: A Simple Solution with Tensorflow Occupancy dynamics helps enable building automation, analysis, and improvement by providing specific information about the number of persons and their locations within a space. Generally, occupancy dynamics involve complicated algorithms and/or complicated equipment. However, it is shown that with a simple sensor network and an elementary Tensorflow neural network, high accuracy results can be achieved. The implementation and results from a tested apartment are given, along with a walk through of the data collection and training of the model. Mark Kurtz has recently graduated from Washington University in St. Louis with my masters in robotics engineering. Before that, Mark graduated with a bachelors in mechanical engineering from Washington University and a bachelors in applied math from Fontbonne University. Over the past 6 years in school he has also worked continuously in data analyst and software engineering positions. Mark currently works for an indoor mapping startup called Aisle411 as a lead software engineer and data analyst. Please confirm your attendance! Due to security concerns we need the First and Last name - as on your government issued ID registered using EVENTBRITE.COM (https://www.eventbrite.com/e/boston-meetup-tensorflow-use-cases-tickets-25018027594)

    1
  • OpenAI & Tensorflow - a Match Made in Heaven?

    PLEASE update your RSVPs at this Eventbrite page (https://www.eventbrite.com/e/boston-meetup-final-rsvp-cross-section-and-deep-dive-into-tensorflow-tickets-22553738840). FYI:for our first meetup, we are doing a joint event together with TensorFlow group (http://www.meetup.com/TensorFlow-Boston/events/226784394/). The meetup will be devoted to TensorFlow solely. Agenda: 5:30 - Doors open. Initiation of the new guild members. 5:45 - Networking. Meet members of the guild. 6:45 - Welcome. Members of the guild to vote on the topics of interests for the future meetings. 7:00 - Fireside chat with Q&Awith Hai Tran, Darshan Patel and Syed Tousif Ahmed 7:15 - Talk #1: Neural Networks with Google TensorFlow 7:45 - Q&A break 8:00 - Talk #2: Transfer learning on Tensorflow in 30 minutes 8:30 - Q&A & wrap-up. Detailed agenda: Fireside chat with Q&A with Hai Tran, Darshan Patel and Syed Tousif Ahmed Talk #1: Neural Networks with Google TensorFlow by Darshan Patel - TensorFlow Terminologies - Softmax regression on MNIST dataset - Convolution Neural Networks(CNN) on MNIST / CIFAR 10 dataset -TensorBoard Graph Visualization -Scikit Flow / Recurrent Neural Networks (RNN) Speaker: Darshan Patel Darshan is a Machine Learning Enthusiastic Second year Computer Science Masters student in the College of Computer and Information science at Northeastern University. Prior to his master study, he was a software developer at TCS, India. His area of interests are Machine/Deep Learning and Artificial Intelligence. Talk #2: "Transfer learning on Tensorflow in 30 minutes"by Syed Tousif Ahmed - What is transfer learning? - How to collect data? - How to retrain a model in tensorflow? - Use cases and application of transfer learning Speaker: Syed Tousif Ahmed Syed is a Computer Vision Developer working at Ahold USA. He is also a student in Computer Engineering at Rochester Institute of Technology. His area of interests are Computer Vision, Machine Learning, Cryptography and Internet of Things.

    12