Anthill Inside 2018
On the current state of academic research, practice and development regarding Deep Learning and Artificial Intelligence.
Accepting submissions
Not accepting submissions
|
DR
Dr. Chiranjiv Roy Industrial Vision & Deep Learning for Manufacturing Quality InspectionQuality Inspection in Manufacturing Industries is of great importance and incurs spends of about $ 3 Billion USD, which is a complex pipeline of tasks which are rigourous, time-consuming and a lot of manual work. Complex testing processes takes weeks to conclude the impact of tests on Parts/Subject where there is a lot of scope for faster inspection using Analytics. The proposed talk elaborates t… more
Technical level: Intermediate
|
|
MG
Mukesh G BigDL: Integrating Deep Learning with Apache SparkBigDL (https://github.com/intel-analytics/BigDL/) is an open source distributed deep Learning library, which is natively integrated with Apache Spark and provides rich deep learning functionalities for Spark. It combines the benefits of “high performance computing” and “Big Data” architecture, so as to provide orders of magnitude speedup than existing out-of-box open source DL frameworks (e.g., C… more
Section: Full talk
Technical level: Beginner
|
|
AS
Aakash N S Deep Learning with High School Math (or Less)You don’t need a PhD or a master’s degree or even a bachelor’s degree in Math/CS to learn and appy deep learning. In most cases, all you need is some programming experience and a quick revision of some high school math e.g. differentiation and matrix multiplication. I’ll show you how you can get up and running in just few hours, and build state of the art deep learning models that solve problems … more
Section: Full talk
Technical level: Beginner
|
|
KC
Karanbir Chahal A Hitchhiker's Guide to Modern Object Detection: A deep learning journey since 2012The ability to detect objects in images has captured the world’s imagination. One can make a decent case that this application is the poster child of deep learning. What really put it on the map. But few people really understands how computers have begun to detect these objects in images with a high accuracy. Which is surprising since it is the backbone of the tech powering self driving cars, dro… more
Section: Full talk
Technical level: Intermediate
|
|
HG
Harsh Gupta  What you cannot do with Machine LearningDuring this “boom” of machine learning and data driven technologies, there is an underlying belief that given enough data any problem is solvable. But like any other technology, machine learning is a tool, appropriate for some problems and not so appropriate for others. Though this talk I would like to remind the community of the things which cannot be done through machine learning. more
Section: Crisp Talk
Technical level: Beginner
|
|
TS
Tapan Shah  A novel Interactive Framework for semi-automated labeling when ground truth resides in free textIn any multi-class supervised learning problem, labeling of training examples is imperative. In most cases, we take expert help in order to execute the annotation, which is time-consuming and often inconsistent. In this talk, we will explain an interactive topic modeling framework to label training examples where the ground truth resides in free text. They key takeaways of this talk will be more
Section: Crisp Talk
Technical level: Intermediate
|
|
SM
Sohan Maheshwar Applying Alexa’s Natural Language To Your ChallengesThis talk will give you a complete picture of all the tools and techniques required to build complex production-quality Alexa skills. You will leave this session knowing how to use Alexa’s dialog management entity resolution and slot elicitation capabilities. The talk will also touch upon some of the key design principles while designing for voice. more
Section: Sponsored talk
Technical level: Intermediate
|
|
KD
Kalpit Desai  The Catalog as a Catalyst - Bringing benefits of Big Data to MSMEsWhile large enterprises have the necessary resources to acquire and process Big Data, the Micro / Small / Medium enterprises in emerging economies like India are far from being ‘data-driven’. This is a huge opportunity untapped, considering that MSMEs account for more than 99% of businesses, and they make up the backbone of our economy. For the opportunity to be leveraged, a crucial pre-requisite… more
Section: Crisp Talk
Technical level: Advanced
|
|
US
Upendra Singh  How organizations can leverage 'Large Scale Graph Based Analytics’ to derive value from their data.An organization’s data is like a living organism - growing, expanding and evolving over time to form complicated and connected systems. This is similar to biological evolution, where life forms evolved from simple unicellular structures to more and more complex multicellular organisms. And as organizations compile more and more data, it is crucial for them to understand that the value of any data… more
Section: Crisp Talk
Technical level: Advanced
|
|
GG
Gaurav Goswami  Adversarial Attacks on Deep LearningDeep learning algorithms are highly popular and being applied to solve various problems with high accuracies. However, they are not infallible and are, in fact, highly susecptible to adversarial attacks. These attacks can manifest in the form of synthetically generated data or perturbed data which mean one thing to a human observer and something completely different to the deep network. Thereby, … more
Section: Crisp Talk
Technical level: Intermediate
|
|
VV
Vikram Vij  Building a next generation Speech & NLU Engine: In pursuit of a Multi-modal experience for BixbyBixby is an intelligent, personalized voice interface for your phone. It lets you seamless switch between voice & type/touch, and supports more than 75 domains (eg. Camera, Gallery, Messages, WhatsApp, Youtube, Uber etc.). It was launched in July 2017 for English and is now available on more than 200 countries with about 8 million registered users. The talk focuses on challenges in deep learning … more
Section: Sponsored talk
Technical level: Beginner
|
|
AN
Aditya Prasad Narisetty Anomaly detection with Variational AutoencodersThere are two types of companies: those that have been hacked, and those who don’t know they have been hacked - John Chambers more
Section: Full talk
Technical level: Beginner
|
|
VG
Vijay Gabale Learning Real-time Object Detection In The Absence of Large-scale DatasetsThis talk will focus an area of computer vision, object detection, that involves automatically localising and classifying different objects from photos and videos. A real-time and accurate object detection technique can help in several critical systems and applications such as self-driving cars (detecting multiple instances of vehicles, humans, etc.), surveillance for public safety, social media … more
Section: Full talk
Technical level: Intermediate
|
|
SS
Saumya Suneja  AI at the Edge: A Software PerspectiveInternet of Things(IoT) devices need to run AI algorithms which are usually associated with high data and computational costs, which is only possible with cloud servers having very powerful systems. Hence these IoT devices aren’t capable of accomplishing much at the ‘edge’ (where the IoT devices are deployed) and their ‘brains’ are located on distant cloud servers. more
Section: Full talk
Technical level: Beginner
|
|
AP
Aditya Patel Looking beyond LSTMs: Alternatives to Time Series Modelling using Neural NetsTime series data, in today’s age, is ubiquitous. With the emerge of sensors, IOT devices it is spanning over all the modern aspects of life from basic household devices to self-driving cars affecting all for lives. Thus classification of time series is of unique importance in current time. With the advent of deep learning techniques , there have been influx of focus on Recurrent Neural Nets (RNN)… more
Section: Crisp Talk
Technical level: Intermediate
|
|
AL
Amar Lalwani Explaining Human Cognition through Deep LearningRevised Bloom’s Taxonomy is very well known and widely used taxonomy for classifying educational objectives. The said taxonomy describes a hierarchical ordering of cognitive skills from simple to complex. The Revised Taxonomy relaxed the strict cumulative hierarchical assumptions of the Original Taxonomy allowing overlaps. We use a knowledge tracing model, Deep Knowledge Tracing (DKT), to investi… more
Section: Crisp Talk
Technical level: Beginner
|
|
AP
Aravind Putrevu Machine Learning and Statistical Methods for Time Series AnalysisIn this talk, I will present a deep algorithmic dive into the new machine learning technologies available in the Elastic Stack and how they can be applied to real datasets. more
Section: Full talk
Technical level: Intermediate
|
|
PN
Puneeth N Building IOT Data pipelines using Prediction IOWe showcase as to why we chose to build ML Pipelines using PredictionIO for our IOT solutions platform. We delve into what PredictionIO is, and how it can help with integrating heterogenous sources of data. more
Section: Crisp Talk
Technical level: Intermediate
|
|
KB
Krishna Bhavsar  Deep Learning - An example implementationIn this talk I intend to showcase one of the problems I solved using Deep Learning framework recently. Resume Classification for a recruitment consultant agency. I shall go through the multiple approaches in which I tried to solve that problem, the obstacles I faced and finally how I came to the final solution. more
Section: Full talk
Technical level: Beginner
|
|
KB
Krishna Bhavsar  Know Your Diabetes Risk - Preventive Health through Risk Prediction and Knowledge BaseIn late 2017 we were wanting to do an academic project in the area of AI/ML. We did a literature review of chronic diseases and found that India is racing to be the Diabetes capital of the world. Then on, the core objective of our study became to predict Diabetes risk of an individual through a prediction module. To augment the prediction module, we trained a text analytics engine with knowledge … more
Section: Crisp Talk
Technical level: Intermediate
|
|
VB
Vineeth N Balasubramanian Going beyond what and asking why: Explainability in Machine/Deep LearningAs machine learning methods get increasingly absorbed in technologies ranging from high-end aerospace systems to low-end consumer technologies, there is a gradual, however steady, increase in the demand for explaining the decisions made by machine learning algorithms. DARPA launched a large initiative in 2016 to further the progress of explainable AI methods, underscoring the need for a concerted… more
Section: Full talk
Technical level: Intermediate
|
|
US
Upendra Singh How organizations can leverage 'Large Scale Graph Based Analytics’ to derive value from their data.An organization’s data is like a living organism - growing, expanding and evolving over time to form complicated and connected systems. This is similar to biological evolution, where life forms evolved from simple unicellular structures to more and more complex multicellular organisms. And as organizations compile more and more data, it is crucial for them to understand that the value of any data… more
Section: Crisp Talk
Technical level: Advanced
|
|
KD
Kalpit Desai The Catalog as a Catalyst - Bringing benefits of Big Data to MSMEsWhile large enterprises have the necessary resources to acquire and process Big Data, the Micro / Small / Medium enterprises in emerging economies like India are far from being ‘data-driven’. This is a huge opportunity untapped, considering that MSMEs account for more than 99% of businesses, and they make up the backbone of our economy. For the opportunity to be leveraged, a crucial pre-requisite… more
Section: Crisp Talk
Technical level: Advanced
|
|
A
Ashwin Attention Mechanisms and Machine ReasoningAttention Mechanisms have been popular for the past couple of years, giving new insights in image and NLP applications using Recurrent Neural Networks. We would like to discuss advances in Attention Mecahnisms in this panel with specific emphasis on two new innovations, Compositional Attention Networks (https://arxiv.org/abs/1803.03067) and Hierarchical Recurrent Attention Networks (https://arxiv… more
Section: Full talk
Technical level: Intermediate
|
|
H(
Hari (ഹരി) Proposing The Sentimental Computer- the Art and Science of Making Computers Understand Sentiment and EmotionCan computers think? Can they have emotion? Such questions are no longer in the realm of fantasy, but are real possibilities in the horizon. Computers are no longer number crunchers; the new age AI make high demands of computers. Deep meaning understanding, sentiment and emotion, translation, inference etc. are the cutting edge technologies that are pushing the frontiers of human computer interac… more
Section: Full talk
Technical level: Advanced
|
|
MG
Madhu Gopinathan  Uncertainty in Deep LearningHow do you deal with uncertainty when making decisions? Presumably, you would collect more information to reduce the uncertainty before making a decision. Now, think about the outputs of deep learning models which can be used to make automated decisions. How will you get uncertainty estimates for these outputs? In this talk, we will focus on quantifying model uncertainty based on recent research … more
Section: Full talk
Technical level: Intermediate
|
|
JD
Jaley Dholakiya Introduction to Game Training using Deep RLFrom AlgphaGo to VizDoom, Deep Reinforcement Learning has revolutionarized the way in which we learn game environments. Especially for people playing CS and Dota in Colleges, having a smarter bot by thier side can help them sleep well, while the bots are fighting against each other. In the talk, we will discussing intuition behind design of Alphago. We will also discuss, what makes Reinforcement … more
Section: Crisp Talk
Technical level: Beginner
|
Anuj Gupta Sarcasm Detection : Achilles Heel of sentiment analysisSentiment analysis has been for long poster boy problem of NLP and has attracted a lot of research. However, despite so much work in this sub area, most sentiment analysis models fail miserably in handling sarcasm. Rise in usage of sentiment models for analysis social data has only exposed this gap further. Owing to the subtilty of language involved, sarcasm detection is a hard problem. more
Section: Full talk
Technical level: Intermediate
|
|
AS
Akhilesh Singh Advances in Deep Learning : Lessons from the fieldDL research has progressed at tremendous speed in recent years. From the demo of Natural conversations at google I/O to BMW X3+ driving itself on road, AI is no more only a topic of research or interest. AI is already everywhere. This talk presents advancements in Deep Learning both in research and field from practitioner’s perspective. This talk has two parts. First part demonstrates the advance… more
Section: Full talk
Technical level: Intermediate
|
|
lT
lavanya TS Product Size Recommendation for Fashion E-commerceRecommending product sizes to customers is an important problem in the e-commerce domain. Though e-commerce is becoming increasingly popular, products such as apparel and shoes remain challenging to buy online and record high return rates. A key customer pain point that leads to excessive product returns is the size-fit problem. This talk (based on the linked WWW 2018 paper) describes some recent… more
Section: Crisp Talk
Technical level: Intermediate
|
|
ND
Nithish Divakar Make your own DL frameworkWe have all used all the high end frameworks that works really well. How about writing a small strip down version of one. In this session, I’ll walk you through how to write a small Deep Learning Framework in Pure python and numpy which has auto grad and optimizers and easy to create models. Also, the framework will be extendible so that you can easily play around with. more
Section: Full talk
Technical level: Intermediate
|
|
H(
Hari (ഹരി) Proposing Building Knowledgeable MachinesKnowledge harvesting from Web-scale text datasets has emerged as an important and active research area over the last decade or so, resulting in the automatic construction of large knowledge graphs (KGs) consisting of millions of entities and relationships among them. This has the potential to revolutionize Artificial Intelligence and intelligent decision making by removing the knowledge bottlenec… more
Section: Full talk
Technical level: Advanced
|
|
H(
Hari (ഹരി) Proposing A very gentle introduction to deep reinforcement learning and applicationsTo be filled Outline To be filled Speaker bio To be filled more
Section: Full talk
Technical level: Advanced
|
|
SK
Shailesh Kumar The evolution in AI thinking and products of the next decadeTo be filled Outline To be filled Speaker bio To be filled more
Section: Full talk
Technical level: Intermediate
|
|
WL
Wei Li Combining Neural Networks and Regression Tree for Dynamic Pricing in Mobile AdvertisingIn a diversified mobile advertising marketplace, it is important to dynamically set the minimum CPM bid in order to maximize revenue as well as meeting CPM expectations. Our approach combines neural networks and a customized regression tree for accurate prediction and effective subsidization. The dNN model predicts revenue and impressions as functions of the minimum CPM bid and other supply/deman… more
Technical level: Advanced
|
|
GS
Gunjan Sharma Neural-network Field Aware Factorisation Machines for Online-behaviour PredictionIn the AdTech mobile-app industry, bidding for each and every ad-request at a suitable price and in real-time is absolutely critical. Thus, there is always a need for more scalable and more accurate prediction models, which drive higher revenue. more
Section: Full talk
Technical level: Advanced
|
|
VP
Vijay Srinivas Agneeswaran, Ph.D  Deep Learning Howlers: Downside of Learning only Statistical RegularitiesIt has been shown in a recent work ( https://arxiv.org/pdf/1711.11561.pdf), that deep convolutional learning networks do not learn higher level abstract concepts, but only statistical regularities. We investigate this claim by taking open source deep learning libraries and testing them out. more
Technical level: Beginner
|
|
HG
Hrishikesh Ganu Building and driving adoption for a robust semantic search systemThis talk focuses on how to use deep learning based sub-word embeddings to create a practical search system robust to queries with mis-spellings, SMS lingo etc. Lifts of upto 20% in search recall compared to commercial solutions were demonstrated with retrieval latency of just 50 milliseconds for queries with mis-spellings and other aberrations. more
Section: Crisp Talk
Technical level: Intermediate
|
|
Anuj Gupta Birds Of Feather (BOF) session: Hubs and spokes of AIIn this session, we will discuss the ‘not-so-glamourous’ aspects of wheel of AI. more
Section: Off The Record session
Technical level: Intermediate
|
|
VG
Vijay Gabale Birds Of Feather (BOF) session: AI and ProductLast 5 years have seen significant progress in many CV, NLP and RL problems. The advent of powerful GPU’s along with insights from last two decades has made inroads into problems that were thought to be unsolvable in the near future. How can these newly solved problems be used for building AI products? What are these problems? How can we convert a Business requirement that requires AI to an AI pr… more
Technical level: Intermediate
|
|
SS
Suchana Seth Birds Of Feather (BOF) session: AI - ethics and privacyIn the aftermath of Cambridge Analytica issue, now public is more conscious about privacy and companies are forced to follow some standards. It also opened a pandora box of issues that has been brewing for a while like bias in algorithms. more
Technical level: Intermediate
|
|
AK
Amit Kapoor Deep Learning in the Browser: Explorable Explanations, Model Inference & Rapid PrototypingWe showcase three live-demos of doing deep learning (DL) in the browser - for building explorable explanations to aid insight, for building model inference applications and even, for rapid prototyping and training ML model - using the emerging client-side Javascript libraries for DL. more
Technical level: Intermediate
|
Hosted by
Anthill Inside is a forum for conversations about risk mitigation and governance in Artificial Intelligence and Deep Learning. AI developers, researchers, startup founders, ethicists, and AI enthusiasts are encouraged to:
more