Deep learning has significantly improved state-of-the-art performance for natural language processing (NLP) tasks, but each one is typically studied in isolation. The Natural Language Decathlon (decaNLP) is a new benchmark for studying general NLP models that can perform a variety of complex, natural language tasks. By requiring a single system to perform ten disparate natural language tasks, decaNLP offers a unique setting for multitask, transfer, and continual learning. decaNLP is maintained by salesforce and is publicly available on github in order to use for tasks like Question Answering, Machine Translation, Summarization, Sentiment Analysis etc.

 
3 favorite thumb_down thumb_up 0 comments visibility_off  Remove from Watchlist visibility  Add to Watchlist
 

Outline/Structure of the Talk

  • Introduction to DecaNLP
  • Objectives
  • Motivation
  • Innovativeness
  • Targeted NLP Tasks
  • Impact
  • Open Source Collaboration on github
  • Patents / Publications in NLP, Computer Vision, AI.

Learning Outcome

People will be able to understand different problems of NLP like:

1. Question Answering

2. Machine Translation

3. Summarization

4. Natural Language Inference

5. Sentiment Analysis

6. Semantic Role Labeling

7. Relation Extraction

8. Goal-Oriented Dialogue

9. Semantic Parsing

10. Commonsense Reasoning

People will know about a unified Framework provided by decaNLP to solve different NLP tasks mentioned above.

Target Audience

People having basic knowledge of NLP, Machine Learning and Deep Learning.

Prerequisites for Attendees

Read basic stuff about NLP, Machine Learning, Deep Learning.

schedule Submitted 3 days ago

Public Feedback

comment Suggest improvements to the Speaker

  • Liked Suvro Shankar Ghosh
    keyboard_arrow_down

    Suvro Shankar Ghosh - Learning Entity embedding’s form Knowledge Graph

    45 Mins
    Case Study
    Intermediate
    • Over a period of time, a lot of Knowledge bases have evolved. A knowledge base is a structured way of storing information, typically in the following form Subject, Predicate, Object
    • Such Knowledge bases are an important resource for question answering and other tasks. But they often suffer from their incompleteness to resemble all the data in the world, and thereby lack of ability to reason over their discrete Entities and their unknown relationships. Here we can introduce an expressive neural tensor network that is suitable for reasoning over known relationships between two entities.
    • With such a model in place, we can ask questions, the model will try to predict the missing data links within the trained model and answer the questions, related to finding similar entities, reasoning over them and predicting various relationship types between two entities, not connected in the Knowledge Graph.
    • Knowledge Graph infoboxes were added to Google's search engine in May 2012

    What is the knowledge graph?

    ▶Knowledge in graph form!

    ▶Captures entities, attributes, and relationships

    More specifically, the “knowledge graph” is a database that collects millions of pieces of data about keywords people frequently search for on the World wide web and the intent behind those keywords, based on the already available content

    ▶In most cases, KGs is based on Semantic Web standards and have been generated by a mixture of automatic extraction from text or structured data, and manual curation work.

    ▶Structured Search & Exploration
    e.g. Google Knowledge Graph, Amazon Product Graph

    ▶Graph Mining & Network Analysis
    e.g. Facebook Entity Graph

    ▶Big Data Integration
    e.g. IBM Watson

    ▶Diffbot, GraphIQ, Maana, ParseHub, Reactor Labs, SpazioDati

  • 45 Mins
    Talk
    Advanced

    Introduction :

    "Special children" includes children who are affected with a complex neuro-behavioral conditions like autism, which includes impairments in social interaction, language development and communication skills, combined with rigid, repetitive behaviors. Children with autism particularly face a very difficult childhood as they have extreme difficulty in communication. They have trouble in understanding what other people think and feel. This makes it very hard for them to express themselves either with words or through gestures.

    Such special children need “special” care for the development of their cognitive abilities. The amount of learning resources required for teaching such children are extremely hard to find and less accessible to many.

    So, can artificial intelligence with the help of modern deep learning algorithms generate animated videos for developing or improving cognitive abilities of such a special group?

    The idea to combat the problem:

    Well, I feel it can be done!

    An animated video consists of 3 main components:

    1. Graphical video (sequence of images put together to tell a story),

    2. A background story and

    3. A relevant background audio or music.

    Now if we have to come up with a system that produces machine generated animated video, we would have to think about these three components:

    1. Machine generated sequence of images with a spatial coherence
    2. Machine generated text, or the story
    3. Machine generated audio or music, that highlights the mood or the theme of the video

    If these three discrete components are put together in a cohesive flow, our purpose can be achieved. And the Deep Learning community has already been able to make significant progress in terms of machine generated images and audio and machine generated text.

    Details about the three pillars of this problem:

    Machine generated sequence of images with a spatial coherence

    Generative Adversarial Networks (GANs) has been quite successful till date to come up with generated images and audio. Also, for our use case, to maintain a coherency in spatial features, Variational Auto Encoders (VAEs) have been even better.

    If we start with a popular use case of a very popular cartoon series, Tom & Jerry, specially modified for autistic children, let’s consider a simple scene where tom is chasing jerry. On an image level, for the entire scene, the posture of tom and jerry will remain constant, only their location will vary in every subsequent image frame in the entire scene. Which means, only their spatial location with respect to the entire image background will vary and hence VAEs will have the potential to implement such a use case as VAEs helps to provide probabilistic descriptions of features or observations in latent spaces.

    Machine generated text, or the story

    Coming to text generation or story generation, recurrent neural networks like Long/Short Term Memory (LSTM) has been quite successful. Already, LSTM has been used to artificially generate chapters from popular novels or stories like Harry Potter and Cinderella. So, for a simple animated video story specially structured for autistic children, LSTM can be effective. Although Gradient Recurrent Units (GRU) can be the other alternative, but till date LSTM has been more successful, so the first preference will always be LSTM.

    Machine generated audio or music

    For music generation, GANs have been proved effective till date. For our use case, Natural Language Processing or NLP can used to determine the type of scene from the generated story, e.g. for the Tom & Jerry scene, it will be a chase scene. Based on this classification, Deep Convolution Generative Adversarial Networks (DCGAN) can be used to generate music which is relevant to such a chase scene and at the same time be soothing and enjoyable to such children!

    Assembling everything together

    Now if we can put all these discrete pieces of the puzzle together, we can come up with a completely machine generated animated video tailor-made for developing and improving cognitive abilities of children with autism. This will be a new progress in the field of Artificial Intelligence!

    These machine generated videos can be trained on Neural Network in such a way that it can be a source of fun and enjoyment for this special group and at the same time reward their good behavior and educate them in a sensitive way without any human dependency.

    Future scope and extension

    As a future scope, if this approach is successful, the gaming industry can adopt usage of such a technology and with the help of reinforcement learning, can come up with machine generated video games and educational games specially designed for such children that can disrupt the entire gaming industry and can be a source of happiness for such children!

  • Liked Joy Mustafi
    keyboard_arrow_down

    Joy Mustafi - Human-Machine Interaction through Multi-Modal Interface with Combination of Speech, Text, Image and Sensor Data

    45 Mins
    Talk
    Intermediate

    Introduction

    In the context of human–computer interaction, a modality is the classification of a single independent channel of sensory input / output between a computer and a human. A system is designated uni-modal if it has only one modality implemented, and multi-modal if it has more than one. When multiple modalities are available for some tasks or aspects of a task, the system is said to have overlapping modalities. If multiple modalities are available for a task, the system is said to have redundant modalities. Multiple modalities can be used in combination to provide complementary methods that may be redundant but convey information more effectively. Modalities can be generally defined in two forms: human-computer and computer-human modalities.

    With the increasing popularity of smartphones, the general public are becoming more comfortable with the more complex modalities. Speech recognition was a major selling point of the iPhone and following Apple products, with the introduction of Siri. This technology gives users an alternative way to communicate with computers when typing is less desirable. However, in a loud environment, the audition modality is not quite effective. This exemplifies how certain modalities have varying strengths depending on the situation. Other complex modalities such as computer vision in the form of Microsoft's Kinect or other similar technologies can make sophisticated tasks easier to communicate to a computer especially in the form of three dimensional movement.

    This talk is based on a physical robot (a personalized humanoid built in MUST Research), equipped with various types of input devices and sensors to allow them to receive information from humans, which are interchangeable and a standardized method of communication with the computer, affording practical adjustments to the user, providing a richer interaction depending on the context, and implementing robust system with features like; keyboard; pointing device; touchscreen; computer vision; speech recognition; motion, orientation etc.

    Cognitive computing makes a new class of problems computable. 

To respond to the fluid nature of users understanding of their problems, the cognitive computing system offers a synthesis not just of information sources but of influences, contexts, and insights. 

These systems differ from current computing applications in that they move beyond tabulating and calculating based on pre-configured rules and programs. 

They can infer and even reason based on broad objectives. In this sense, cognitive computing is a new type of computing with the goal of more accurate models of how the human brain or mind senses, reasons, and responds to stimulus. 

It is a field of study which studies how to create computers and computer software that are capable of intelligent behavior. This field is interdisciplinary, in which a number of sciences and professions converge, including computer science, electronics, mathematics, statistics, psychology, linguistics, philosophy, neuroscience and biology.

    Computer–Human Modalities

    Computers utilize a wide range of technologies to communicate and send information to humans:

    • Vision – computer graphics typically through a screen
    • Audition – various audio outputs
    • Tactition – vibrations or other movement
    • Gustation (taste)
    • Olfaction (smell)
    • Thermoception (heat)
    • Nociception (pain)
    • Equilibrioception (balance)

    Human–computer Modalities

    Computers can be equipped with various types of input devices and sensors to allow them to receive information from humans. Common input devices are often interchangeable if they have a standardized method of communication with the computer and afford practical adjustments to the user. Certain modalities can provide a richer interaction depending on the context, and having options for implementation allows for more robust systems.

    • Keyboard
    • Pointing device
    • Touchscreen
    • Computer vision
    • Speech recognition
    • Motion
    • Orientation

    Project Features

    Adaptive: They MUST learn as information changes, and as goals and requirements evolve. They MUST resolve ambiguity and tolerate unpredictability. They MUST be engineered to feed on dynamic data in real time.

    Interactive: They MUST interact easily with users so that those users can define their needs comfortably. They MUST interact with other processors, devices, services, as well as with people.

    Iterative and Stateful: They MUST aid in defining a problem by asking questions or finding additional source input if a problem statement is ambiguous or incomplete. They MUST remember previous interactions in a process and return information that is suitable for the specific application at that point in time.

    Contextual: They MUST understand, identify, and extract contextual elements such as meaning, syntax, time, location, appropriate domain, regulation, user profile, process, task and goal. They may draw on multiple sources of information, including both structured and unstructured digital information, as well as sensory inputs (visual, gestural, auditory, or sensor-provided).

  • Liked Dr. Saptarsi Goswami
    keyboard_arrow_down

    Dr. Saptarsi Goswami - Meta features and clustering based approaches for feature selection

    45 Mins
    Tutorial
    Beginner

    Feature selection is one of the most important processes for pattern recognition, machine learning and data mining problems. A successful feature selection method facilitates improvement of learning model performance and interpretability as well as reduces computational cost of the classifier by dimensionality reduction of the data. Feature selection refers to the retention of discriminatory features while discarding the redundant and irrelevant features. In this process, a subset of D features are selected from a set of N features (D<N). There is another way of achieving dimensionality reduction by projecting higher dimensional data to lower dimension, normally referred to feature extraction. This thesis refers to the former one i.e. feature subset selection. Optimal feature subset selection method comprises of developing 1) an evaluation function for measuring the goodness of a feature or a feature subset and 2) a search algorithm to find out the best subset of features from all possible subsets of the whole feature set. Based on the nature of the objective function used in the search algorithms, feature subset selection algorithms are broadly classified into filter approach and wrapper approach. Classifier dependent wrapper approaches use classifier accuracy as the objective function while filter approaches use any evaluation function representing the intrinsic characteristics of the data set and the resulting feature subset works equally well for any classifier. This work focusses on filter based feature subset selection approach. In this work, initially a study has been done with currently available search based filter type feature selection algorithms for supervised as well as unsupervised classification with both the single objective and multi-objective evaluation functions. Some improvements over the current algorithms have been proposed and their efficiency has been examined by simulation experiments with bench mark data sets. In the second step, an inexpensive feature evaluation measure based on feature relevance to be used with a filter type feature selection for unsupervised classification has been proposed. It has been noticed during literature study that the concept of feature relevance in case of unsupervised classification is difficult to form and current methods are complex and time consuming. The proposed measure which considers individual variability as well as overall variability of the dataset,is found to be effective compared to the current methods by simulation experiments with bench mark data sets. Thirdly, it seems that the most of the current feature selection algorithms are based on search strategies to find out the best feature subset from the available feature set. For a large number of features, exhaustive search is computationally prohibitive which leads to combinatorial optimization problem and some sort of heuristic is used for the solution. With the increase of the number of features, the computational time for optimal feature subset selection increases.An alternative solution to this problem is to use clustering of the features to find out the best feature subset which is not yet explored sufficiently. In this work, an efficient clustering based feature selection algorithm has been proposed and simulation experiments have been done with bench mark data sets. The main contributions of the proposed algorithm are introduction of a novel method to determine the optimal number of clusters, a way of interpretation of the importance of the feature clusters and a method of selection of the final subset of features from the feature clusters. Finally, it is found that though lots of feature selection algorithms are available, it is very difficult to decide which algorithm is suitable for a particular real world application. Here a study has been done to establish the relation between the feature selection algorithm and the characteristics of the data set. A technique has been proposed to define a data set according to its intrinsic characteristics represented by some meta-features. Finally a feature selection strategy is recommended based on the characteristics of the data set and has been implemented with bench mark data sets to judge its effectiveness.

  • Liked Siboli mukherjee
    keyboard_arrow_down

    Siboli mukherjee - Real time Anomaly Detection in Network KPI using Time Series

    20 Mins
    Experience Report
    Intermediate

    Abstract:

    How to accurately detect Key Performance Indicator (KPI) anomalies is a critical issue in cellular network management. In this talk I shall introduce CNR(Cellular Network Regression) a unified performance anomaly detection framework for KPI time-series data. CNR realizes simple statistical modelling and machine-learning-based regression for anomaly detection; in particular, it specifically takes into account seasonality and trend components as well as supports automated prediction model retraining based on prior detection results. I demonstrate here how CNR detects two types of anomalies of practical interest, namely sudden drops and correlation changes, based on a large-scale real-world KPI dataset collected from a metropolitan LTE network. I explore various prediction algorithms and feature selection strategies, and provide insights into how regression analysis can make automated and accurate KPI anomaly detection viable.

    Index Terms—anomaly detection, NPAR (Network Performance Analysis)

    1. INTRODUCTION

    The continuing advances of cellular network technologies make high-speed mobile Internet access a norm. However, cellular networks are large and complex by nature, and hence production cellular networks often suffer from performance degradations or failures due to various reasons, such as back- ground interference, power outages, malfunctions of network elements, and cable disconnections. It is thus critical for network administrators to detect and respond to performance anomalies of cellular networks in real time, so as to maintain network dependability and improve subscriber service quality. To pinpoint performance issues in cellular networks, a common practice adopted by network administrators is to monitor a diverse set of Key Performance Indicators (KPIs), which provide time-series data measurements that quantify specific performance aspects of network elements and resource usage. The main task of network administrators is to identify any KPI anomalies, which refer to unexpected patterns that occur at a single time instant or over a prolonged time period.

    Today’s network diagnosis still mostly relies on domain experts to manually configure anomaly detection rules such a practice is error-prone, labour intensive, and inflexible. Recent studies propose to use (supervised) machine learning for anomaly detection in cellular networks . ellular networks, a common practice adopted by network administrators is to monitor a diverse set of Key Performance Indicators (KPIs), which provide time-series data measurements that quantify specific performance aspects of network elements and resource usage. The main task of network administrators is to identify any KPI anomalies, which refer to unexpected patterns that occur at a single time instant or over a prolonged time period.

    Today’s network diagnosis still mostly relies on domain experts to manually configure anomaly detection rules such a practice is error-prone, labour intensive, and inflexible. Recent studies propose to use (supervised) machine learning for anomaly detection in cellular networks .

  • Liked Ishita Mathur
    keyboard_arrow_down

    Ishita Mathur - How GO-FOOD built a Query Semantics Engine to help you find the food you want to order

    Ishita Mathur
    Ishita Mathur
    Data Scientist
    GO-JEK Tech
    schedule 1 week ago
    Sold Out!
    45 Mins
    Case Study
    Beginner

    Context: The Search problem

    GOJEK is a SuperApp: 19+ apps within an umbrella app. One of these is GO-FOOD, the first food delivery service in Indonesia and the largest food delivery service in Southeast Asia. There are over 300 thousand restaurants on the platform with a total of over 16 million dishes between them.

    Over two-thirds of those who order food online using GO-FOOD do so by utilising text search. Search engines are so essential to our everyday digital experience that we don’t think twice when using them anymore. Search engines involve two primary tasks: retrieval of documents and ranking them in order of relevance. While improving that ranking is an extremely important part of improving the search experience, actually understanding that query helps give the searcher exactly what they’re looking for. This talk will show you what we are doing to make it easy for users to find what they want.

    GO-FOOD uses the ElasticSearch stack with restaurant and dish indexes to search for what the user types. However, this results in only exact text matches and at most, fuzzy matches. We wanted to create a holistic search experience that not only personalised search results, but also retrieved restaurants and dishes that were more relevant to what the user was looking for. This is being done by not only taking advantage of ElasticSearch features, but also developing a Query semantics engine.

    Query Understanding: What & Why

    This is where Query Understanding comes into the picture: it’s about using NLP to correctly identify the search intent behind the query and return more relevant search results, it’s about the interpretation process even before the results are even retrieved and ranked. The semantic neighbours of the query itself become the focus of the search process: after all, if I don’t understand what you’re trying to ask for, how will I give you what you want?

    In the duration of this talk, you will learn about how we are taking advantage of word embeddings to build a Query Understanding Engine that is holistically designed to make the customer’s experience as smooth as possible. I will go over the techniques we used to build each component of the engine, the data and algorithmic challenges we faced and how we solved each problem we came across.

  • Liked Siboli mukherjee
    keyboard_arrow_down

    Siboli mukherjee - AI in Telecommunication -An Obstacle or Opportunity

    45 Mins
    Talk
    Executive

    Introduction

    “Alexa, launch Netflix!”

    No longer limited to providing basic phone and Internet service, the telecom industry is at the epicentre of technological growth, led by its mobile and broadband services in the Internet of Things (IoT) era.This growth is expected to continue,The driver for this growth? Artificial intelligence (AI).

    Artificial Intelligent applications are revolutionizing the way telecoms operate, optimize and provide service to their customers

    Today’s communications service providers (CSPs) face increasing customer demands for higher quality services and better customer experiences (CX). Telecoms are addressing these opportunities by leveraging the vast amounts of data collected over the years from their massive customer base. This data is culled from devices, networks, mobile applications, geolocations, detailed customer profiles, services usage and billing data.

    Telecoms are harnessing the power of AI to process and analyse these huge volumes of Big Data in order to extract actionable insights to provide better customer experiences, improve operations, and increase revenue through new products and services.

    With Gartner forecasting that 20.4 billion connected devices will be in use worldwide by 2020, more and more CSPs are jumping on the bandwagon, recognizing the value of artificial intelligence applications in the telecommunications industry.

    Forward-thinking CSPs have focused their efforts on four main areas where AI has already made significant inroads in delivering tangible business results: Network optimization, preventive maintenance, Virtual Assistants, and robotic process automation (RPA)

    Network optimisation

    AI is essential for helping CSPs build self-optimizing networks (SONs), where operators have the ability to automatically optimize network quality based on traffic information by region and time zone. Artificial intelligence applications in the telecommunications industry use advanced algorithms to look for patterns within the data, enabling telecoms to both detect and predict network anomalies, and allowing operators to proactively fix problems before customers are negatively impacted.

    Some popular AI solutions for telecoms are ZeroStack’s ZBrain Cloud Management, which analyses private cloud telemetry storage and use for improved capacity planning, upgrades and general management; Aria Networks, an AI-based network optimization solution that counts a growing number of Tier-1 telecom companies as customers, and Sedona Systems’ NetFusion, which optimizes the routing of traffic and speed delivery of 5G-enabled services like AR/VR. Nokia launched its own machine learning-based AVA platform, a cloud-based network management solution to better manage capacity planning, and to predict service degradations on cell sites up to seven days in advance.

    Predictive maintenance

    AI-driven predictive analytics are helping telecoms provide better services by utilizing data, sophisticated algorithms and machine learning techniques to predict future results based on historical data. This means telecoms can use data-driven insights to can monitor the state of equipment, predict failure based on patterns, and proactively fix problems with communications hardware, such as cell towers, power lines, data centre servers, and even set-top boxes in customers’ homes.

    In the short-term, network automation and intelligence will enable better root cause analysis and prediction of issues. Long term, these technologies will underpin more strategic goals, such as creating new customer experiences and dealing efficiently with business demands. An innovative solution by AT&Tis using AI to support its maintenance procedures: the company is testing a drone to expand its LTE network coverage and to utilize the analysis of video data captured by drones for tech support and infrastructure maintenance of its cell towers.Preventive maintenance is not only effective on the network side, but on the customer’s side as well.Dutch telecom KPN analyses the notes generated by its call centre agents, and uses the insights generated to make changes to the interactive voice response (IVR) system.

    Virtual Assistants

    Conversational AI platforms — known as virtual assistants — have learned to automate and scale one-on-one conversations so efficiently that they are projected to cut business expenses by as much as $8 billion in the next five years. Telecoms have turned to virtual assistants to help contend with the massive number of support requests for installation, set up, troubleshooting and maintenance, which often overwhelm customer support centre Using AI, telecoms can implement self-service capabilities that instruct customers how to install and operate their own devices.

    Vodafone introduced its new chatbot — TOBi to handle a range of customer service-type questions. The chatbotscales responses to simple customer queries, thereby delivering the speed that customers demand. Nokia’s virtual assistant MIKA suggests solutions for network issues, leading to a 20% to 40% improvement in first-time resolution.

    Robotic process automation (RPA)

    CSPs all have vast numbers of customers and an endless volume of daily transactions, each susceptible to human error. Robotic Process Automation (RPA) is a form of business process automation technology based on AI. RPA can bring greater efficiency to telecommunications functions by allowing telecoms to more easily manage their back office operations and the large volumes of repetitive and rules-based processes. By streamlining execution of once complex, labor-intensive and time-consuming processes such as billing, data entry, workforce management and order fulfillment, RPA frees CSP staff for higher value-add work.

    According to a survey by Deloitte, 40% of Telecom, Media and Tech executives say they have garnered “substantial” benefits from cognitive technologies, with 25% having invested $10 million or more. More than three-quarters expect cognitive computing to “substantially transform” their companies within the next three years.

    Summary

    Artificial intelligence applications in the telecommunications industry is increasingly helping CSPs manage, optimize and maintain not only their infrastructure, but their customer support operations as well. Network optimization, predictive maintenance, virtual assistants and RPA are examples of use cases where AI has impacted the telecom industry, delivering an enhanced CX and added value for the enterprise overall.

  • Liked Debjyoti Paul
    keyboard_arrow_down

    Debjyoti Paul - Transfer Learning in Unsupervised text processing

    45 Mins
    Tutorial
    Advanced

    Today we are facing enormous amount of unstructured textual data. Given a text processing problem, how to start? What models to build language model with? Can models trained in similar domains be exploited. These are some trailing questions.

    1. When and how to use Transfer Learning- new vocabulary? 2. Challenges in Text processing and Transfer Learning 3. Effectively method selection for transfer learning 4. Applications 5. How to validate your model?

    Presentation on Aspect detection in unsupervised domain using Transfer Learning from structure prediction.

  • Liked Kaushik Dey
    keyboard_arrow_down

    Kaushik Dey - Managed Algorithms at Edge leveraging decentralized learning

    45 Mins
    Talk
    Advanced

    The problem of network behavior prediction has been an ongoing study by researchers for quite a while now. Network behavior typically exhibits a complex sequential pattern and is often difficult to predict. Nowadays there are several techniques to predict the degradation in Network KPIs like throughput, latency etc., using various machine learning techniques like Deep Neural Networks, where the initial layers have learnt to map the raw features like performance counter measurements, weather, system configuration details etc into a feature space where classification by the final layers can be performed.

    Given the initial number of counters( which constitutes the dimensions) is substantial (more than 2000 in number) the problem requires huge amount of data to train the Deep Neural Networks. Often this needs resources and time and more importantly this requires provisioning of huge amount of data for every trial. Given each node generates huge amount of data ( data on every 2000 counters generated at 15 minutes interval for each of 6 cells in an eNodeB) and the data needs to be transported across several hundred of eNodeBs to one central data center, it requires a very fat data pipe and consequently huge investment to enable a predictive fault prediction apparatus across the network.

    The alernative is to have a compute infrastructure at the node and take the intelligence at the edge. However the challenge is given the huge amount of data generated in a single node having a compute at each node was proving to be expensive. Nowadays this compute requirement at node could be reduced through use of transfer learning. However the other challenge is on sharing the intelligence and developing a system which is collectively intelligent across nodes.

    Network topology, climate features and user patterns vary across regions and service providers and hence an unique model is often necesarry to serve the node. However in order to deal with unseen patterns intelligence from other nodes can be useful which leads us to building an global model which again leads to the challenge of fat data pipeline requirement which makes it commercially less attractive.

    In order to get around this challenge, an combination of federated learning is used in combination with transfer learning.

    This presentation details such deep learning architectures which combines federated learning with transfer learning to enable construction and updation of Global models which imbibes intelligence from nodes but can be constructed by a consensus mechanism whereby weights and changes to weights of local models are shared to global. Also the local models are periodically updated once global model update iteration is complete. Further updation of local models is only done in final layers and initial layers are freezed. This reduces the compute requirement at node also...

    The above principles are being implemented as First of a kind implementation and has prooved to be a success across multiple customers in delivering a compelling ML enabled fault prediction and self-healing mechanism but keeping the investments in infrastructure lower than would have been required in traditional Deep Learning architectures

    This talk will specifically detail the leverage of above principles of federated and transfer learning on LSTMs..