ADCOM 2018 Programme
21 September 2018, Friday [Day 1] |
||
09:00 AM | Tutorial Registration | |
09:30 AM | Dr. Ashish Tendulkar, Google TensorFlow is the second machine learning framework that Google created and used to design, build, and train deep learning models.You can use the TensorFlow library do to numerical computations, which in itself doesn’t seem all too special, but these computations are done with data flow graphs. In these graphs, nodes represent mathematical operations, while the edges represent the data, which usually are multidimensional data arrays or tensors, that are communicated between these edges. Tensorflow Workshop Content |
|
01:30 PM | LUNCH | |
Inaugural Session of ADCOM 2018 |
||
---|---|---|
02:00 PM | ADCOM 2018 Registration | |
03:00 PM | Welcome | Saragur M. Srinidhi, President, ACCS |
03:05 PM | Conference Chairs | Dr. Balaraman Ravindran, IIT Madras & Mr. K.R. Sanjiv, CTO Wipro Ltd |
03:20 PM | Technical Program Chair | Dr. Ramasuri Narayanam, IBM Research |
03:30 PM | Inaugural Address | Dr. Ajay Kumar, GoI |
04:15 PM | Conferment of ACCS – CDAC Foundation Award by Dr. S. Sadagopan, IIITB on Prof. V. Kamakoti, IIT Madras & Prof. P. Vijay Kumar, IISc |
|
04:45 PM | ACCS – CDAC Foundation Lecture Prof. V. Kamakoti, IIT Madras SHAKTI and FinLibre – Towards self-sufficiency in compute technologies for an emerging Digital IndiaWith the advent of Digital India the need for a family of computing devices within the country is projected to see an exponential increase. This ranges from small processor cores for tiny embedded devices to large high performance processor cores for transactional and cloud based systems. The RISE Lab, Department of Computer Science and Engineering, IIT Madras conceived of the SHAKTI project in 2011 with a target to deliver indigenous commercial grade processors for our country by 2020. This talk shall outline the journey over the last 7 years and the future roadmap of the project that is driven by Digital India initiatives. The talk shall also briefly touch upon certain open source initiatives by The RISE Lab in the area of data analytics. |
|
05:30 PM | High Tea & Networking | |
06:30 PM | Close | |
22 September 2018, Saturday [Day 2] |
||
08:30 AM | ADCOM 2018 Registration | |
09:30 AM | ACCS – CDAC Foundation Lecture Prof. P. Vijay Kumar, IISc Erasure Coding for Big Data To ensure reliability within a data center, information pertaining to a file is stored in distributed and redundant fashion across a network of storage nodes. The amount of data housed can run into the petabytes, thereby causing the industry to increasingly turn towards erasure-correcting codes to ensure that this reliability comes with the smallest possible storage overhead. Given that the the failure of an individual storage unit is a common occurrence, there is urgent need for erasure-coding techniques that are efficient in dealing with node failure. In part response to this challenge, coding theorists have come up with two new classes of codes, namely Regenerating Codes and Codes with Locality. This talk will provide an accessible overview of these exciting recent developments in coding theory. Several classes of codes will be presented, many of which originated within the authors research group. |
|
10:20 AM | Keynote – Dr. P. Anandan, Wadhwani AI | |
11:10 AM | AI Innovation Jam Interlude – Ramesh Sundararaman | |
11:25 AM | TEA-BREAK | |
TRACK 1: Machine Learning for Healthcare (Ramanath Padmanabhan, ARM) | TRACK 2: Social Media and Text Understanding (Prof. Chandrashekar Ramanathan, IIITB) | |
11:45 AM | Optimised Deep Net Architectures for ECG Analysis; Nirmala B, Priyanka P and Raghuram S – Ramaiah Institute of Technology |
Fuzzy Approach towards English-Hindi Translation; Shashi Pal Singh, Ajai Kumar and Shubhi Vaish – CDAC, Pune |
12:10 PM | Drowsiness detection by analysis of EEG signal with the help of Machine Learning; Venkata Phanikrishna B, Suchismita Chinara and Mahasweta Sarkar – National Institute of Technology, Rourkela |
Recurrent Neural Network Architectures with Trained Document Embeddings for Flagging Cyber-Aggressive Comments on Social Media; Shylaja S S, Abhishek Narayanan, Abhijith Venugopal and Abhishek Prasad – PES University, Bangalore |
12:35 PM | H. U. B – Eye, Hearing Using Bone Conduction and Seeing through Deep Neural Networks; Akshath Varugeese, Anuj Nawal, Harsh Mehta and Sasipriya P – Vellore Institute of Technology, Chennai |
System for Semi-Automatic Chatbots Query Classification Training Corpus Generation; Udit Chandna and Dr. Manjunath Ramachandra Iyer – Wipro Ltd. |
01:00 PM | LUNCH | |
02:00 PM | Keynote – Udayan Ganguly, IIT Bombay Spiking Neural Networks (SNN) aspire to replicate brain-like computing. They are different from Artificial Neural Networks (ANN) a significant way i.e. the token in information exchange in the generation, communication and reception of voltage spikes in the network – akin to biology. The learning rules are also biology inspired – which depend on “local information” i.e. information that a neuron can observe around it (i.e. its own spike and that of its neighbors). SNNs do not require a global error calculation like in back propagation in ANNs – when the error of the result of the network needs to be calculated. Thus, SNNs aim to model biology to mimic biological energy efficiency and exceptional performance in hardware. In this talk, we will explore algorithms, circuits and devices from our research group that offer solutions to hardware implementations of biology. |
|
02:50 PM | AI Innovation Jam Interlude – Jailendra Kumar | |
TRACK 3: PhD Forum – Chair – Dr. N. Ramamurthy, ACCS | TRACK 4: Image Understanding and AI (Hariharan Natarajan, Wipro Ltd.) | |
03:00 PM | PhD Forum 2018 – An Information-theoretic Approach to User Privacy: A Single Database Private Information Retrieval; Radhakrishna Bhat and N R Sunitha – Siddaganga Institute of Technology, Tumkur |
High-Resolution Satellite Image Compression using Uncertain Color Space Based Method; Pritpal Singh and Kinjal Rabadiya – CMPICA, CHARUSAT, Anand, Gujarat |
03:15 PM | PhD Forum 2018 – Replica Update Technique in RDRTDBS: Issues & Challenges; Pratik Shrivastava and Udai Shanker – Madan Mohan Malaviya Unversity of Technology, Gorakhpur |
Single window platform for Explainable AI; Chandrashekar B Nagaraj, Manjunath Ramachandra Iyer, Sanjib Khetan, Vinutha B N and Pallavi Ramesh Naik – Wipro Ltd. |
03:30 PM | PhD Forum 2018 – Residue-to-Binary converters for the moduli set {2n-1-1, 2n+k, 2n-1}; Madhavilatha Mvn, Rashmi Ramesh Racch and Ananda Mohan Pemmaraju Venkata – CDAC Benguluru |
Real-Time Headgear Detection in Videos Using Deep Learning Based Feature Extraction with A Supervised Classifier; Rajdeep Pal, Ranjana Seshadri, Swarnashree Mysore Sathyendra and Natarajan S – PES Institute of Technology |
03:45 PM | PhD Forum 2018 – Investigation of Bio-inspired Algorithms in Localization of Sensor Nodes; Vaishali Kulkarni and Veena Desai- M.S.Ramaiah University of Applied Sciences |
|
04:00 PM | Ph.DForum 2018- Design and parametric Study on Printed MIMO Antenna for Wirless Applications; Shiddanagouda F Byanigoudra, Dr.Vani R.M and Dr.P.V. Hunagund – Gulbarga University Kalaburagi |
|
04:15 PM | BREAK | |
04:40 PM | Keynote – Dr. Manish Gupta, Videoken Inc Rapid changes enabled by technology are making it an imperative for virtually every human being to commit to lifelong learning. The growth of online courses has led to the widespread availability of high quality video lectures on practically any topic. However, much of formal learning, in educational institutions as well as corporations, makes inadequate use of these resources, and purely online learning continues to face challenges that limit its effectiveness. We describe our attempts to improve learning via a platform called VideoKen. Our platform uses novel techniques to support search and recommendation of educational videos on a given topic. It applies machine learning techniques to index videos, inspired by the analogy with text books, by automatically generating a table of contents and a glossary, and to gain insights from the learner’s interactions with the videos. We describe the need to support social learning at multiple levels, both to enable faculty to conveniently share their curated video content within and across institutes, as well as learners to curate specific video clips and share with their classmates, friends or the organization at large. We describe many outstanding challenges in creating an engaging and personalized learning experience for each learner, and describe our preliminary efforts to deal with those challenges. |
|
05:30 PM | AI Applications Innovation JAM | |
23 September 2018, Sunday [Day 3] |
||
09:30 AM | Keynote – Dr. Sriram Rajamani, Microsoft Research Will cover work done at Microsoft Research India in the areas of AI, ML and IoT. Cover topics ranging from theory to practice ranging from data science theory, to low resource NLP, low resource ML, IoT systems, ML-systems interplay and AI/Ml solutions for societal scale problems. |
|
10:20 AM | Keynote – Dr. Shailesh Kumar, Reliance Jio | |
11:10 AM | AI Innovation Jam Interlude | |
11:25 AM | BREAK | |
TRACK 5: Deep Learning and Applications (Dr. Ram Rustagi, KSIT) | TRACK 6: PhD Forum – Chair – Dr. N. Ramamurthy, ACCS | |
11:45 AM | Robotic Grasper based on an End-to-End Neural Architecture using Raspberry Pi; Pranav Sreedhar and Suja Palaniswamy – Amrita School of Engineering, Bengaluru |
(11:45 AM ) Incremental Mining Algorithms: Adapting to Dynamic Data; Panthadeep Bhattacharjee and Pinaki Mitra- Indian Institute of Technology Guwahati |
12:10 PM | Relation Extraction using Deep Learning; Sriram Chaudhury, Amrit Bhaskar and Arkaprava Nayek- Wipro Ltd. | (12:00 PM )PhD Forum 2018 – Design and Implementation of Object Detection, Tracking, Counting and Classification Algorithms using Artificial Intelligence for Automated Video Surveillance Applications; Mohana Mohana and Ravish Aradhya H V – R.V.College of Engineering, Bengaluru |
12:35 PM | Towards a Hybrid Model for CPU Usage Prediction of Smartphone Users; Sriram Sankaran and Manish Gupta – Amrita University |
(12:15 PM )Weakly Supervised Recurrent Neural Network for Video Segmentation; Aditya K and Raghavendra V. Kulkarni – M S Ramaiah Univeristy of Applied Sciences |
(12:30 PM )Perceptual QoE Modeling and Optimization for Adaptive Video Streaming; Nagabhushan Eswara, Sumohana Channappayya, Abhinav Kumar and Kiran Kuchi – Indian Institute of Technology Hyderabad |
||
01:00 PM | LUNCH | |
02:00 PM | AI Application Innovation JAM | |
04:00 PM | VALEDICTORY |