SIGML

SIGML - Special Interest Group on Machine Learning of Computer Science , was born out of an effort to bring together people interested in areas of Machine Learning, Data Mining and related fields.

It is an endeavor to bring people who share an excitement in Machine Learning, Computer Vision, NLP and Data Mining to discuss latest developments and research options. The group aims at organizing problem solving sessions, seminars, research days, workshops and guest lectures. For more details view WikiPage

This webpage is dedicated to documenting the progress of the group and will include abstracts of lectures delivered by SIGML speakers and additional references. SIGML forum is envisaged to promote exchange of ideas and encourage research in emerging areas related to Machine Learning. Researchers, practitioners and interested students are invited to participate in this forum.

SIGML Lecture Series

This semester SIGML will be holding lectures on extreme multi-label classification & deep learning theory . Later it would be holding lectures related to Bayesian Model and Online & Stochastic Optimization and Learning .

SIGML Reading Groups

We are planning to form reading groups and decided to meet on regular basis to discuss and solve some problems. This semester our objective is to get some exposure to Deep Learning , Kernels and Bayesian Graphical Model. Recently we also started the Cognitive Science Group. For subscribing to deep learning group, send a mail to one of the two co-ordinators or sigml [at] cse.iitk.ac.in . For other groups , we will be updating soon!!.

SIGML Seminar Series

SIGML would continue to hold seminars this semester on various areas of Machine Learning and Data Mining.

If you wish to be added to the mailing list you can subscribe by filling the following Form .Outside IIT Kanpur personal can also subscribe subject to approval by filling the Form .If you have any other queries regarding SIGML, please contact at sigml [at] cse.iitk.ac.in
Professors

Purushottam Kar (CSE Department)

Online and streaming methods in learning and optimization , Kernel methods & Learning theory

Piyush Rai (CSE Department)


Machine Learning, Statistics, Artificial Intelligence

Amitabha Mukerjee (CSE Department)

Cognitive Science, Machine Learning, Artificial Intelligence Vision, Language and Robotics

Vinay P. Namboodiri (CSE Department)

Computer vision, machine learning, computer graphics, image processing

Gaurav Sharma (CSE Department)

Computer Vision and Machine Learning

Harish Karnick (CSE Department)

Automated and Commonsense Reasoning, AI, Programming Languages.

Arnab Bhattacharya (CSE Department)

Databases, Data Mining, Bioinformatics

Medha Atre (CSE Department)

Graph databases, Distributed Data Management and Query Optimization

Tanaya Guha (EE Department)

Multimedia Signal Processing, Affective Computing, Computer Vision, Image/Video Analyses

Gaurav Pandey (EE Department)

Computer Vision, Mobile Robotics, Image Processing, Machine Learning, Sensor Data Fusion, Information Theory
Students
Vivek Gupta Deep Learning for NLP, Vector Embedding, Cross View Learning, Computer Vision , Kernels
Chirag Gupta Learning Theory, Online Learning, SVMs, structured losses, Causality, Optimization
Avikalp Kumar Gupta NLP, Neural Networks, Applications of Machine Learning
Unnat Jain Optical character recognition, Place detection, Alignment of images
Amartya Sanyal High Dimensional Regression, Deep Learning, Sparse Regression, Econometric models
M.Arunothia Image Processing
Akshat Agarwal SLT, Generalization ability, VC dimension, classifier combination, computer stereo vision, visual odometry
Ayush Mittal Deep Neural Networks, Domain Adaptation/Transfer Learning, Active Learning, Natural Language Processing
Adarsh Chauhan Active learning(used in CNNs), Application of CNNs for image classification, Deep learning, computer vision.
Ayush Sekhari Learning theory, Gaussian processes, Approximate inference
Viveka Kulharia Kernels, Semantics
Drishti Wali Neural Networks and theoretical aspects of Machine Learning
Yeshi Dolma Active learning(used in CNNs), Application of CNNs for image classification, Deep learning, computer vision.
Kriti Joshi Machine Learning, Computer Vision
Rahul Kumar Sevakula Machine Learning, Classification algorithms, Health Monitoring of machines, Object matching
Sharbatanu Chatterjee ML, Computational Neuroscience, AI
Avi Singh Computer Vision, Robotics
Abhimanyu Goyal Document/Multi document/query based Summarisation
Siddhant Manocha Systems for machine learning , computer vision
Anurendra Kumar Machine learning for Multimedia, Bayesian Inference, Deep learning
Subhash Chandra Tiwari Detecting and Decoding vehicle license plate in Surveillance video
Mirza Mazhar Ali Beg Dataset labeling via crowd sourcing
Aditya Modi Learning Theory, SVM's, Semisupervised Learning, Online Learning, Optimisation
K.Vinay Sameer Raja Computer Vision and NLP, Image based Q-A systems etc , Phrase and paragraph vectors.
Arnab Ghosh Deep Learning applications to NLP and Computer Vision
Amlan Kar Computer Vision, Multimodal Learning, Deep Learning
Vasu Sharma Deep Learning, Audio and Signal Processing, Computer Vision , Convolutional Neural Nets for Image Recognition
Aayush Mudgal Machine Learning, AI in Education, Computer Vision
Kundan Kumar Machine Learning, NLP , Computer Vision
Arpit Agarwal Localization methods with Deep learning and object recognition during SLAM
Satyam Shivam Computer Vision, Depth Estimation from Single Monocular Image, Spatial Text Tagging, Robot Motion Planning
Aishwarya Jadhav Neural nets, machine learning, computer vision

Background Courses

These courses form the basis of many ideas in Machine Learning. They form part of compulsary undergraduate curriculum at IIT Kanpur and you'll anyways be doing them. Thus, this is a guide on what you should be looking to take away from these courses, if you're interested in Machine Learning. If you are not from the CSE department, some of them might not be compulsary.

Course No.: Title Department Semester Comments
MTH101 : Introduction to Real Analysis MTH Even
MSO201 / ESO 209 : Probability and Statistics MTH Even
MTH102 : Introduction to Linear Algebra MTH Even
MTH203 : Ordinary and Partial Differential Equation MTH Even
ESC101 , CS210 , CS345 : Programming and Algorithms courses CSE Even
CS201 : Discrete Mathematics CSE Even

Intermediate Courses

One of these should be taken as soon as possible, if you're interested in working in Machine Learning. Since you'll only get an opportunity to take them as electives in the 5th or 6th semester, you can replace them with various MOOCs offered on edX and coursera. (See section on online courses below)

Course No.: Title Department Semester Comments
CS671 : Natural Language Processing CSE Even Current Semester (Prof. Harish Karnick)
CS771 : Machine learning: tools, techniques and applications CSE Even
CS365 : Artificial Intelligence CSE Even
CS315 : Principles of Database Systems CSE Even

Advanced Courses

If you enjoyed your introductory courses on Machine Learning, and are aching to delve head on into the field, take any of these courses. The focus is often on contemporary work, projects, and presentations. Most of these courses involve a decent research component, and hence are often heavy on mathematics.

Course No.: Title Department Semester Comments
CS772 : Skypline Queries in Databases CSE ODD Current Semester (Prof Arnab Bhattarcharya)
CS774 : Optimization Techniques CSE ODD Current Semester (Prof Purushottam Kar)
CS771 : Machine Learning CSE ODD Current Semester (Prof Piyush Rai)
CS698N : Recent Adavances in Computer Vision CSE ODD Current Semester (Prof Gaurav Sharma)
CS719 : Data Streaming: Algorithms And Systems CSE ODD Current Semester (Prof Sumit Ganguly)
CS698F : Advanced Data Management CSE ODD Current Semester (Prof Megha Atre)
CS773 : Online Learning and Optimization CSE Even
SE367 : Introduction to Cognitive Science CSE Even
CS772 : Probabilistic Machine Learning CSE Even
EE698 : Probabilistic Mobile Robotic EE Even
EE609 : Convex Optimization in SP/COM EE Even
CS638 : Formal Methods for Robotics and Automation EE Even
CS676 : Computer Vision and Image Processing CSE Even
CS781 : Cognition: Memory CSE Odd
CS782 : Cognitive Semantics CSE Even
CS672 : Natural Language Processing Semantics CSE Odd
CS673 : Machine Translation CSE Even
CS674 : Knowledge Discovery CSE Even
CS677 : Data and Information Fusion CSE Odd
CS678 : Learning with Kernels CSE Even
CS679 : Machine Learning for Computer Vision CSE Even
CS719 : Data Streaming Algorithms and Systems CSE Odd
CS685 : Data Mining CSE Even
CS686 : Biometric Recognition CSE Odd
CS726 : Topics in Multimedia CSE Even
CS789 : Special Topics in Language Acquisition and Origins CSE Odd
CS718 : Sublinear Algorithms for Processing Massive Data Sets CSE Odd

Other Relavent Courses/Topics

In addition to the courses listed above, there are a variety of other courses in CSE and other departments at IITK (EE and MTH) that you may wish to explore.

Course No.: Title Department Semester Comments
EE604 : Image Processing EE Even
EE602 : Statistical Signal Processing EE Even
EE601 : Mathematical Methods in Signal Processing EE Even
EE627 : Speech Signal Processing EE Even
EE626 : Topics in Stochastic Processes EE Even
EE671 : Neural Networks EE Even
EE672 : Computer Vision and Document Processing EE Even
EE677 : Knowledge Based Man Machine Systems EE Even
EE678 : Neural Systems and Networks EE Even


Machine Learning Tools

Deep Learning

Data Science Programme

Miscellaneous

Machine Learning Research Day(MLRD)

Talks by Professors and Eminent Researchers

Talks by Research Students

Nov 26, 2015- Seminar Talk: Multimedia Analytics for Videos

Speaker: Dr Arjit Biswas, Xerox Research Lab , India
Time: Nov 26, 2015 , 3:30 PM - 5:00 PM
Location: RM 101
Host: Dr. Vinay Namboodari

First, Talk about a distance learning method in non-vector spaces, where the triangle inequality is used to propagate the pairwise constraints to the unsupervised image pairs. This approach can work with any pairwise distance and does not require any vector representation of images.Second,novel approach to jointly segment and classify egocentric/first-person activity videos of daily-life. First, ego-centric activity classifiers are learnt in a novel multiple instance learning (MIL) based framework, which can remove distractors present in long and complex egocentric-activities. Second, these classifiers are used in a dynamic programming framework to simultaneously segment an egocentric video into individual activities and classify them.
[ Details ]

Nov 23, 2015- Seminar Talk: Expanded Parts Model for Human Analysis and Nonlinear Models for Classification and Embeddings

Speaker: Dr Gaurav Sharma, Max Planck Institute for Informatics, Germany
Time: Nov 23, 2015 , 3:30 PM - 5:00 PM
Location: RM 101
Host: Dr. Vinay Namboodari

First part of the talk I will introduce our Expanded Parts Model (EPM) for recognizing human attributes (e.g. young, short hair, wearing suit) and actions (e.g. running, jumping) in still images.An EPM is a collection of part templates which are learnt discriminatively to explain specific scale-space regions in the images (in human centric coordinates).
I will introduce our novel approach for learning nonlinear support vector machine (SVM) corresponding to commonly used kernels in computer vision
[ Details ]

October 14, 2015- Seminar Talk: Neural Attention Models for Natural Language Grounding and Generation

Speaker: Mohit Bansal, Toyota Technological Institute at Chicago
Time: October 14, 2015 , 4:00 PM - 5:30 PM
Location: RM 101
Host: Dr. Purushottam Kar

Neural sequence-to-sequence, encoder-decoder models have recently shown strong promise in the areas of machine translation and image captioning as end-to-end models that require little domain-specific knowledge or resources. Incorporating an attention or alignment step into this encoder-decoder architecture helps further by learning to focus on parts of the input sequence that are salient for generating a particular step in the output sequence.
[ Details ]

October 14, 2015- Seminar Talk: Machine Learning for Multimedia Analytics

Speaker: Dr OM Deshmukh, Xerox Research Lab , India
Time: October 14, 2015 , 6:00 PM - 7:30 PM
Location: RM 101
Host: Dr. Vinay Namboodari

The amount of multimedia data available online continues to grow exponentially. The recent surge in video-capturing devices has led to thousands of hours of video being captured every day. Automatic analysis of this large amount of content is a challenging research problem but has a wide range of applications, e.g.: efficient navigation, saliency detection, deep semantic analysis for summarization, event recognition etc.
[ Details ]


Jan 4,2016 - Seminar Talk: Topics in Nonnegative Matrix Factorization:

Speaker: Abhishek Kumar (IBM TJ Watson Research Center)
Time: Jan 4, 2016 , 3:00 PM - 4:30 PM
Location: RM 101
Host: Dr. Piyush Rai

The goal in nonnegative matrix factorization (NMF) is to express, exactly or approximately, a given matrix as a product of two nonnegative matrices of smaller inner dimension. NMF and its variants have been widely used for extracting interpretable features and patterns in various applications, including text, vision, and speech. Computing NMF has been shown to be NP-hard. In the first part, I will talk about fast conical hull algorithms for NMF under the so-called separability assumption which makes the NMF problem tractable. I will show applications of separable NMF to the problem of Video foreground-background separation, comparing it with Robust PCA which is a widely used method for this problem. In the second part, talk on a related problem of Semi-nonnegative matrix factorization where only one of the factors is constrained to be nonnegative. I will talk about conditions for tractability, and exact and heuristic algorithms for computing Semi-NMF. .
[ Details ]

May 5, 2015- [Skype Seminar] : Introduction to Dialogue Systems (Cortana)

Speaker: Puneet Agarwal (MS IDC Cortana Team)
Time: May 5, 2016 , 5:00 PM - 6:00 PM
Location: KD 101
Host: Vivek Gupta

Dialogue Systems are taking over the world, the trend is so strong that they will become part of our life just like Search engine did many years back. In this interactive talk we will look at this trend, what is causing this huge momentum, and what are various systems that are coming up. We will briefly also talk about some of the high level components.
[ Details ]

May 4, 2015- [Skype Seminar] : Bridge Correlational Neural Networks for Multilingual Multimodal Representation Learning

Speaker: Dr. Balaraman Ravindran (IIT Madras)
Time: May 4, 2016 , 3:30 PM - 5:00 PM
Location: KD 101
Host: Vivek Gupta

Recently there has been a lot of interest in learning common representations for multiple views of data. These views could belong to different modalities or languages. Typically, such common representations are learned using a parallel corpus between the two views (say, 1M images and their English captions). In this work, we address a real-world scenario where no direct parallel data is available between two views of interest (say, V1 and V2) but parallel data is available between each of these views and a pivot view (V3). We propose a model for learning a common representation for V1, V2 and V3 using only the parallel data available between V1V3 and V2V3. The proposed model is generic and even works when there are n views of interest and only one pivot view which acts as a bridge between them.
[ Details ]

Apr 7,2016 - Online Learning Algorithms at Scale: Applications to Search and Sponsored Search

Speaker: Rahul Agrawal
(Principal Researcher
,Microsoft India R&D Private Limited)
Time: Apr 7, 2016 , 5:00 PM
Location: KD 102
Host: Dr. Purushottam Kar

Rahul Agrawal is a Principal Researcher leading a team of applied scientists in Microsoft India R&D Private Limited Bangalore, India, where he primarily works in the domain of large scale machine learning to understand user intent and show relevant ads on Bing. Prior to this, he was with Yahoo Labs, where he was involved in click prediction models for display advertising. He has completed his Masters in Computer Science from IISc Bangalore in 2004. His research interests are in the areas of large scale machine learning, information retrieval, spectral graph clustering and text mining.
[ Details ]

Mar 15,2016 -[Skype Seminar] Lecture on Entity Mining

Speaker: Dr. Manish Gupta
(Senior Applied Scientist , Bing Team , MS IDC)
Time: Mar 15, 2016 , 3:30 PM - 5:00 PM
Location: KD 101
Host: Vivek Gupta

Entity mining is a hot area of research. At Microsoft Bing, we perform a large number of entity mining tasks which continuously populate and use Bing's knowledge graph, Satori. The talk will be in two parts. The first part will discuss a few entity mining tasks and their solutions: (1) entity synonym discovery, (2) entity attribute discovery and augmentation, (3) entity linking, (4) entity acronym expansion, (5) entity conflation, and (6) entity actions. The second part will discuss how Microsoft applied various entity mining algorithms for building the following applications: (1) entity linking in the Microsoft Edge and Snapshots on Tap, (2) extracting fictional character entities from books, (3) extracting disaster event entities from Twitter, and (4) event entity linking for sports events.
[ Details ]

March 11,2016 - [Skype Talk] Object Detection in Presence of Hard Examples

Speaker: Subhabrata Debnath
Time: 11 March 2016 , 5:00 PM - 6:00 PM
Location: KD 101
Host: Vivek Gupta
Object detection involves finding the location of the object of interest in an image. This is done by learning a detector using training images containing the location of the object as input. Thus the robustness of the detector directly depends on the quality of training set. In our work we try to achieve robust detection even in the presence of visually hard images in the training data. In our setting, we are presented with labels which indicate the presence or absence of an object in an image but not their explicit locations. This is called a weakly supervised setting. We aim to learn a detector which can classify a test image as well as find the location of the bounding box containing the object in the image. This can be done by modeling the location of the object as a latent parameter and learning both the location and the classifier jointly during training. We show how using a variation of Outliers Robust-SVM and Self paced learning with latent variables can be used to obtain good results in this scenario. We show our results on three classes of the Pascal Voc 2007 dataset and present a comparison with existing methods


[ Details ]

Mar 4,2016 - [Skype Seminar]: Teaching Machine - Next Fortier of AI

Speaker: Dr Shailesh Kumar ,Co-founder at Third Leap
Time: Mar 4, 2016 , 5:30 PM - 7:10 PM
Location: KD 101
Host: Vivek Gupta

The Idea of Education - What we learn, How we learn, and Why we learn - is going through a fundamental transformation today. The confluence of the ubiquitously available Internet, proliferation of devices to access the Internet, and progress in Artificial Intelligence and Machine Learning are going to power this impending paradigm shift in the way humanity teaches its children.
[ Details ]

Mar 4,2016 - [Skype Seminar] Tutorial on Structural Output Prediction

Speaker: Nitish Gupta ,Phd Student UUIC
Time: Mar 2, 2016 , 8:00 PM - 9:00 PM
Location: KD 101
Host: Vivek Gupta

Learning functional dependencies between arbitrary input and output spaces especially in problems involving multiple dependent output variables and structured output spaces is extremely difficult and cannot be achieved using trivial supervised learning algorithms for multi-class classification. In this talk, which will be more of a tutorial, I will start by giving a brief introduction to supervised methods for binary classification using linear classifiers and extending this idea to Multi-class classification. The focus in Multi-class classification will be on One vs. All, All vs. All, Multi-class SVM and Constraint Classification approaches. I will then introduce the problem of structured output prediction and present the various challenges it poses in training and inference. I will conclude the talk with a brief tutorial on a widely used supervised learning approach called the Structured SVM.
[ Details ]

Feb 8,2016 - [Skype Talk] Introduction and Survey to 3D Vision :

Speaker: Shubham Tulsiani, University of California, Berkeley
Time: Feb 8, 2016 , 6:50 PM
Location: KD101
Host: Vivek Gupta

Abstract: This talk will serve as an introduction and survey for 3D vision - the task of developing a 3D visual understanding from 2D images. We will look at various aspects of this problem and review recent approaches. The talk will focus on explaining the core ideas, outlining the learning formulations and highlighting the common elements of the approaches involved.
[ Details ]
[ Slides]

Jan 27,2016 - [Skype Talk] Generative adversarial networks Unsupervised Learning (Deep Learning) :

Speaker: Soumith Chintala,Researcher Facebook AI Research
Time: Jan 27, 2016 , 6:50 PM
Location: KD101
Host: Vivek Gupta

Abstract: In this talk, we will discuss recent advances in method of neural network optimization called adversarial networks, and their application to generative modeling.Specifically, we will look at the domains of image generation / synthesis -where the goal is to​​ generate fake images that are imperceptible from the real image distribution. We will finally look into using GANs as an unsupervised learning method.
[ Details ]

Jan 21,2016 - [Skype Talk] Understanding Word Embedding :

Speaker: Omer Levy,Bar-Ilan University,NLP group
Time: Jan 21, 2016 , 6:50 PM
Location: KD101
Host: Vivek Gupta

Abstract: Neural word embeddings, such as word2vec (Mikolov et al., 2013), have become increasingly popular in both academic and industrial NLP. These methods attempt to capture the semantic meanings of words by processing huge unlabeled corpora with methods inspired by neural networks and the recent onset of Deep Learning. The result is a vectorial representation of every word in a low-dimensional continuous space. These word vectors exhibit interesting arithmetic properties (e.g. king - man + woman = queen) (Mikolov et al., 2013), and seemingly outperform traditional vector-space models of meaning inspired by Harris's Distributional Hypothesis (Baroni et al., 2014). Our work attempts to demystify word embeddings, and understand what makes them so much better than traditional methods at capturing semantic properties.
[ Details ]

Jan 10,2016 - Few selected problems in image processing and computer vision :

Speaker: Prof. Simant Dube, New College of Florida
Time: Jan 15, 2016 , 3:00 PM
Location: KD103
Host: Dr. Arnab Bhattacharya

Abstract:
Last couple of decades have witnessed amazing progress in image processing and computer vision. We take a journey through this evolving landscape, visiting the following problems:
1) increasing the resolution of a digital image
2) image classification of protein crystal images
3) image classification of everyday objects
We also discuss how big data computing, large scale machine learning and computer vision are converging these days to create new technologies and to push the cutting edge further.
[ Details ]

[Cancel] Jan 10,2016 - Seminar Talk: An Introduction to Deep Learning :

Speaker: Prof. Lawrence Carin (Duke University)
Time: Jan 10, 2016 , 2:00 PM - 3:00 PM
Location: RM 101
Host: Dr. Piyush Rai

This talk will introduce Deep Learning from the perspective of generative statistical models, and factor analysis. We will see how the idea of sparsity in such models has a direct counterpoint in sigmoid belief networks, and other related models. The talk will explain how single-layer models of this type may naturally be extended to "deep" multi-layered settings. It will be demonstrated that for many applications (e.g., image and video analysis) convolutional factor models can be convenient, for which the convolutional neural networks have conventionally been used as a natural tool. We will finally see the usefulness of these deep models with a diverse set of applications.
[ Details ]

Jan 5, 2016- Seminar Talk: Incorporating Structure for Natural Language Understanding

Speaker: Snigdha Chaturvedi, U Maryland, College Park
Time: Jan 5, 2016 , 3:45 PM - 5:00 PM
Location: KD 102
Host: Dr. Purushottam Kar

This talk emphasizes the use of structured approaches towards addressing Natural Language Understanding (NLU) problems. We argue that many NLU tasks can benefit from using models that are capable of incorporating not just linguistic cues, but also the contexts in which these cues appear. In this talk, we present a structured approach to model the 'flow of information.in text to solve two seemingly distinct problems: (i) Identifying need for instructor intervention in MOOC discussion forums, and (ii) Analyzing a paragraph to identify if a desire expressed in it was fulfilled.
[ Details ]

February 12,2017 - Generative Adversarial Network

Speaker: Ian Goodfellow
Time: 11 am
Venue: KD 101
Host: SIGML

We are having an interaction session with Ian Goodfellow on the 12th of February(Sunday) on the topic of GANs(Generative Adversarial Network) and adversarial learning. Ian will begin the session by talking about some problems in the domain, which are interdisciplinary in nature and would hopefully be interesting to people, who want to partake in conducting research in this domain.
[ Details ]

February 10,2017 - How Deep Learning Revolutionized Speech Recognition

Speaker: Sunayana Sitaram
Time: 5:00 PM - 6:00 PM
Venue: KD 101
Host: SIGML

Designing of general-purpose learning algorithms is a long-standing goal of artificial intelligence. A general purpose AI agent should be able to have a memory that it can store and retrieve information from. Despite the success of deep learning in particular with the introduction of LSTMs and GRUs to this area, there are still a set of complex tasks that can be challenging for conventional neural networks. Those tasks often require a neural network to be equipped with an explicit, external memory in which a larger, potentially unbounded, set of facts need to be stored. They include but are not limited to, reasoning, planning, episodic question-answering and learning compact algorithms. To view the complete publication list and speaker profile, please visit: http://sarathchandar.in
[ Details ]

January 21,2017 - Towards Principled Methods for Training Generative Adversarial Networks

Speaker: Martin Arjovsky
Time: 21st January 2017, 7pm pm
Venue: KD 101
Host: Vivek Gupta

We have recently seen a couple of breakthroughs in speech recognition – Microsoft’s systems have reached human parity in transcribing speech and Baidu’s system is said to be 3 times faster than human transcription for typing text messages, and just as accurate as humans. Does this mean that the problem is solved? How did Deep Learning manage to revolutionize the field? What does it take to replicate this success in new languages and domains? In this talk, I will present the problem of Automatic Speech Recognition and talk about the various Machine Learning solutions that have been proposed over the years. We will look at research and results on specific datasets and trace the improvement of speech recognition systems on them.
[ Details ]

January 20,2017 - [Skype Seminar] Memory Augmented Neural Networks

Speaker: Sarath Chandar
Time: 20th January 2017, 5:30 pm
Venue: KD 101
Host: Vivek Gupta

Designing of general-purpose learning algorithms is a long-standing goal of artificial intelligence. A general purpose AI agent should be able to have a memory that it can store and retrieve information from. Despite the success of deep learning in particular with the introduction of LSTMs and GRUs to this area, there are still a set of complex tasks that can be challenging for conventional neural networks. Those tasks often require a neural network to be equipped with an explicit, external memory in which a larger, potentially unbounded, set of facts need to be stored. They include but are not limited to, reasoning, planning, episodic question-answering and learning compact algorithms. To view the complete publication list and speaker profile, please visit: http://sarathchandar.in
[ Details ]

January 13,2017 - [Skype Seminar] Learning with Complex Performance Metrics

Speaker: Nagarajan Natarajan
Time: 13th January 2017, 5 pm
Venue: RM 101
Host: Vivek Gupta

Prediction tasks arising in modern day recommender systems often necessitate complex performance metrics for evaluation. For instance, classification accuracy (or the “0-1 loss”) metric is ill-suited for rare event classification problems such as medical diagnosis, fraud detection, click rate prediction and text retrieval applications. Practitioners instead employ alternative metrics better tuned to imbalanced classification, such as the F-measure. An important theoretical question concerning complex metrics is characterizing their optimal decision functions given the inherent uncertainty in the data and the labeling process.
[ Details ]

January 10,2017 - Intuitive Physics and Intuitive Behavior

Speaker: Pulkit Agrawal
Time: 10th January 2017, 3 pm
Venue: KD 101, CSE Department
Host: Vinay P. Namboodiri

The mammalian brain inspired the architecture of neural networks that power state of the art visual recognition systems. Is the only similarity between the brain and the neural networks architectural? I will show that a deep neural network trained for object recognition mimics the hierarchy of representations in the human visual cortex (http://arxiv.org/abs/1407.5104). This leads to an interesting hypothesis that building a neural network based system for performing visuomotor tasks may provide new tools for studying the neural mechanisms that integrate sensory and motor processing. However, while superhuman performance has been achieved on specific visuomotor tasks (such as ATARI games), a system that can perform a general set of day to day object manipulation tasks is yet to be built.
[ Details ]

September 27,2016 - CNN-based Single Image Obstacle Avoidance on a Quadrotor

Speaker: Dr. Prunarjay Chakravarty
Time: Sept 27, 2016 , 17:00 IST
Location: RM101
Host: Vinay P. Namboodiri

This talk will be about the use of a single forward facing camera for obstacle avoidance on a quadrotor. A ConvolutionalNeural Network(CNN) is trained for estimating depth from a single image. The depth map is then fed to a behaviour arbitration based control algorithm that steers the quadrotor away from obstacles. Experiments conducted demonstrate the use of single image depth for controlling the quadrotor in both simulated and real environments.
[ Details ]

September 7,2016 - [Skype Talk] Weakly Supervised Object Detection

Speaker: Dr. Hakan Bilen
Time: Sept 7, 2016 , 19:30 IST
Location: KD101
Host: Gaurav Sharma

Weakly supervised learning of object detection is an important problem in image understanding that still does not have a satisfactory solution. In this talk, we address this problem by improving different aspects of the standard multiple instance learning based object detection. We first present a method that can represent and exploit presence of multiple object instances in an image. Second we further improve this method by imposing similarity among objects of the same class. Finally we propose a weakly supervised deep detection architecture that can exploit the power of deep convolutional neural networks pre-trained on large-scale image-level classification tasks
[ Details ]

August 29,2016 - Deep Learning based Large Scale Visual Recommendation and Visual Semantic Embedding for E-Commerce

Speaker: Dr. Krishnendu Chaudhury
Time: Aug 29, 2016 , 17:00
Location: RM 101
Host: Harish Karnick

Recommending catalog items that are visually similar to a catalog item the user is browsing is an important problem in e-commerce. We refer to this problem as "CIVR" (Catalog Image based Visual Recommendation). CIVR is a very challenging task in its own right, due to the extreme variety among the catalog items that could be deemed similar - dress items maybe hanging or laid flat on a table or worn by different models or mannequins having different complexions and/or hair color, standing in different poses etc. Furthermore, the human notion of similarity is extremely abstract and complex. Two t-shirts, one with a batman print and another with a superman print, maybe called "similar" by human beings, while, in terms of pixels, there maybe very little in common between the two images.
[ Details ]

August 28,2016 - Machine Learning Research Day

Speaker: Click Here
Time: Aug 10, 2016 , 10 am onwards
Location: RM 101
Host: SIGML Team

We are organising the SIGML *MLRD(Machine Learning Research Day)* on *28th of September*. It is meant to be a platform for showcasing your somewhat recent work via a presentation or a poster. This can be something you have done in your internship, UGP, as a hobby, a thesis or some other project. The main motivation is to get to know what kind of work is being done by other people in this institute related to the field of Machine Learning and at the same time increasing interactions between the faculty and the students. For some of us it might just be getting some exposure in the field of ML by attending it.
[ Details ]

Overview

Following are recent research topics explored in the group for more details please refer to the involved faculty homepages.

  • Machine Learning in Computer Vision
  • Machine Learning in Text Mining
  • Data analytics
  • Data mining and knowledge discovery
  • AI for Robotics
  • MultiMedia Analysis
  • Optmizations
  • Information extraction
  • Support Vector Machines and Kernel based Learning Methods
  • Structured Prediction
  • Latent and structural SVM
  • Graphical Models
  • Robust Learning Methods for Uncertain Data
  • Statistical Learning Theory, Statistical Consistency of Learning Algorithms
  • Deep Learning for NLP
  • Binary Classification in Presenece of severse label Imbalance
  • High dimensional regression, sparse recovery , compressive sensing
  • Robust regression in the presence of corruptions
  • Online and stochastic optmiization and learning
  • Reinforcement learning
  • Introduction to statistical Learning theory
  • Neural Autoencoder RNN and LSTM
  • Active learning
  • Online Learning
  • Domain Adaption
  • Deep classificaltion web heirarchy
  • Word-Vector Embeddings
  • Performance Metrics
  • Probabilistic modeling and Bayesian learning
  • Approximate Bayesian inference
  • Nonparametric Bayesian learning

Depending on the interest of the participating audience other topics can also be explored. The interest shall however be to discuss recent papers and ideas in the machine learning community.

March 10, 2015 - Seminar Talk: Efficient Contextual Semi-bandits

Speaker: Akshay Krishnamurthy, Carnegie Mellon University
Time: March 10, 2015, 2:00pm-3:00pm
Location: GDC 5.516
Host: Inderjit S. Dhillon
In the contextual bandit problem, a learner, for a series of rounds, observes a context, makes an action, and receives reward for this action, but does not see the potential reward for other possible actions. Consequently, the learner must negotiate an exploration-exploitation tradeoff, whereby it explores to understand good actions for each context, but also exploits so that it achieves high reward over the course of the interaction. In contextual bandit problems with large action spaces, existing algorithms fail to achieve competitive statistical performance because exploring spaces is prohibitively expensive.

March 10, 2015 - Seminar Talk: Efficient Contextual Semi-bandits

Speaker: Akshay Krishnamurthy, Carnegie Mellon University
Time: March 10, 2015, 2:00pm-3:00pm
Location: GDC 5.516
Host: Inderjit S. Dhillon
In the contextual bandit problem, a learner, for a series of rounds, observes a context, makes an action, and receives reward for this action, but does not see the potential reward for other possible actions. Consequently, the learner must negotiate an exploration-exploitation tradeoff, whereby it explores to understand good actions for each context, but also exploits so that it achieves high reward over the course of the interaction. In contextual bandit problems with large action spaces, existing algorithms fail to achieve competitive statistical performance because exploring spaces is prohibitively expensive.
Conference Title : ICML 2015
Publication Title : Optimizing Non-decomposable Performance Measures: A Tale of Two Classes
Topic : Learning Theory
Authors: Harikrishna Narasimhan, Purushottam Kar and Prateek Jain
Conference Website : http://icml.cc/2015/
Publication Link : http://www.cse.iitk.ac.in/users/purushot/papers/sgd-tpr-tnr.pdf
Conference Title : ICML 2015
Publication Title : Surrogate Functions for Maximizing Precision at the Top
Topic : Learning Theory
Authors: Purushottam Kar, Harikrishna Narasimhan, and Prateek Jain
Conference Website : http://icml.cc/2015/
Publication Link : http://www.cse.iitk.ac.in/users/purushot/papers/preck.pdf
Conference Title : AAAI 2017
Publication Title : Scalable Optimization of Multivariate Performance Measures in Multi-instance Multi-label Learning
Topic : Optimization
Authors: Apoorv Aggarwal, Sandip Ghoshal, Ankith M S, Suhit Sinha, Ganesh Ramakrishnan, Purushottam Kar, and Prateek Jain
Conference Website : http://www.aaai.org/Conferences/AAAI/aaai17.php
Publication Link : http://www.cse.iitk.ac.in/users/purushot/papers/miml-perf.pdf
Conference Title : ICDM 2016
Publication Title : Optimizing the Multiclass F-measure via Biconcave Programming
Topic : Optimization
Authors: Weiwei Pan, Harikrishna Narasimhan, Purushottam Kar, Pavlos Protopapas, and Harish G. Ramaswamy
Conference Website : http://icdm2016.eurecat.org/
Publication Link : http://www.cse.iitk.ac.in/users/purushot/papers/macrof1.pdf
Conference Title : KDD 2016
Publication Title : Stochastic Optimization Techniques for Quantification Performance Measures
Topic : Optimization
Authors: Purushottam Kar, Shuai Li, Harikrishna Narasimhan, Sanjay Chawla, and Fabrizio Sebastiani
Conference Website : www.kdd.org/kdd2016/
Publication Link : http://www.cse.iitk.ac.in/users/purushot/papers/quant.pdf
Conference Title : NIPS 2015
Publication Title : Sparse Local Embeddings for Extreme Multi-label Classification
Topic : Optimization
Authors: Kush Bhatia, Himanshu Jain, Purushottam Kar, Manik Varma, and Prateek Jain
Conference Website : https://nips.cc/Conferences/2015
Publication Link : http://www.cse.iitk.ac.in/users/purushot/papers/x1.pdf
Conference Title : NIPS 2015
Publication Title : Robust Regression via Hard Thresholding
Topic : Optimization
Authors: Kush Bhatia, Prateek Jain, and Purushottam Kar
Conference Website : https://nips.cc/Conferences/2015
Publication Link : http://www.cse.iitk.ac.in/users/purushot/papers/rr-torrent.pdf
Conference Title : AAAI 2017
Publication Title : Nonnegative Inductive Matrix Completion for Discrete Dyadic Data
Topic : Probabilistic Machine Learning
Authors: Piyush Rai
Conference Website : http://www.aaai.org/Conferences/AAAI/aaai17.php
Publication Link :
Conference Title : AISTATS 2016
Publication Title : Topic-Based Embeddings for Learning from Large Knowledge Graphs
Topic : Probabilistic Machine Learning
Authors: Piyush Rai , Changwei Hu , Lawrence Carin
Conference Website : http://www.jmlr.org/proceedings/papers/v51/
Publication Link : http://www.cse.iitk.ac.in/users/piyush/papers/topic_kg_aistats.pdf
Conference Title : AISTATS 2016
Publication Title : Non-negative Matrix Factorization for Discrete Data with Hierarchical Side-Information
Topic : Probabilistic Machine Learning
Authors: Piyush Rai , Changwei Hu , Lawrence Carin
Conference Website : http://www.jmlr.org/proceedings/papers/v51/
Publication Link : http://www.cse.iitk.ac.in/users/piyush/papers/nmf_hier.pdf
Conference Title : ECML 2016
Publication Title : Deep Distance Metric Learning with Data Summarization
Topic : Probabilistic Machine Learning
Authors: Piyush Rai,Wenlin Wang, Changyou Chen, Wenlin Chen and Lawence Carin
Conference Website : http://www.ecmlpkdd2016.org/
Publication Link : http://www.cse.iitk.ac.in/users/piyush/papers/dSNC.pdf
Conference Title : NIPS 2015
Publication Title : Large-Scale Bayesian Multi-Label Learning via Topic-Based Label Embeddings
Topic : Probabilistic Machine Learning
Authors: Piyush Rai , Changwei Hu , Lawrence Carin , Ricardo Henao
Conference Website : https://nips.cc/
Publication Link : http://people.duke.edu/~pr73/recent/rai15bmlpl.pdf
Conference Title : Computer Graphics Forum (Journal) 2016
Publication Title : Sketchsoup: Exploratory Ideation using Design Sketches
Topic : Computer Graphics
Authors: Rahul Arora (University of Toronto/IIT Kanpur/Inria Sophia-Antipolis); Ishan Darolia (IIT Kanpur); Vinay P. Namboodiri (IIT Kanpur); Karan Singh (University of Toronto); Adrien Bousseau (Inria Sophia-Antipolis)
Conference Website : http://onlinelibrary.wiley.com/journal/10.1111/(ISSN)1467-8659
Publication Link : http://www.cse.iitk.ac.in/users/vinaypn/papers/CGF16/sketchinterp.pdf
Conference Title : AAAI 2017
Publication Title : Contextual RNN-GANs for Abstract Reasoning Diagram Generation
Topic : Computer Vision
Authors: Arnab Ghosh, Viveka Kulharia, Amitabha Mukerjee, Vinay Namboodiri, Mohit Bansal
Conference Website : http://www.aaai.org/home.html
Publication Link : https://arxiv.org/abs/1609.09444
Conference Title : ICASSP, 2017
Publication Title :On Random Weights for Texture Generation in One Layer Neural Networks
Topic : Computer Vision
Authors: Mihir Mongia, Kundan Kumar, Akram Erraqabi, Yoshua Bengio
Conference Website : https://arxiv.org/abs/1612.06070
Publication Link : https://arxiv.org/pdf/1612.06070v1.pdf
Conference Title : ACCV 2016
Publication Title : Using Gaussian Processes to Improve Zero-Shot Learning with Relative Attributes
Topic : Computer Vision
Authors: Yeshi Dolma, Vinay Namboodari
Conference Website : http://http//www.accv2016.org/
Publication Link :
Conference Title : ECCV | TASK-CV 2016
Publication Title : Deep Attributes for One-Shot Face Recognition
Topic : Computer Vision
Authors: Aishwarya Jadhav, Vinay Namboodari
Conference Website : http://adas.cvc.uab.es/task-cv2016/
Publication Link :
Conference Title : INFOCOM 2015
Publication Title : Trajectory Aware Macro-cell Planning for Mobile Users
Topic : Databases
Authors: Shubhadip Mitra, Sayan Ranu, Vinay Kolar, Arnab Bhattacharya, Ravi Kokku, Aditya Telang, Sriram Raghavan
Conference Website :
Publication Link :
Conference Title : BIBM 2016
Publication Title : Assisting User to Achieve Optimal Sleep Using Ambient
Topic : Bioinformatics
Authors: Vivek Gupta, Siddhant Mittal, Sandip Bhaumik, Raj Roy
Conference Website : https://cci.drexel.edu/ieeebibm/bibm2016/
Publication Link :
Conference Title : ACCS 2015
Publication Title : The effects of category entropy, defining feature saliency and response time limit on category learning
Topic : Cognitive Science
Authors: Sujith Thomas and Harish Karnick
Conference Website :
Publication Link :
Conference Title : CEUR-WS
Publication Title : Using Attentive Focus to Discover Action Ontologies from Perception
Topic : Cognitive Science
Authors: Amitabha Mukerjee
Conference Website : http://ceur-ws.org/Vol-481/
Publication Link :
Conference Title : CEUR-WS
Publication Title : Symbol Emergence in Design
Topic : Cognitive Science
Authors: Amitabha Mukerjee, Madan Dabbeeru
Conference Website : http://ceur-ws.org/Vol-481/
Publication Link :
Conference Title : CogSci 2016
Publication Title : Factors Influencing Categorization Strategy in Visual Category Learning
Topic : Computer Vision
Authors: Sujith Thomas and Harish Karnick
Conference Website :
Publication Link :
Conference Title : CVPR 2016
Publication Title : {CP-mtML}: {C}oupled Projection multi-task Metric Learning for Large Scale Face Retrieval
Topic : Computer Vision
Authors: Karan Sikka and Gaurav Sharma and Marian Bartlett
Conference Website : http://cvpr2016.thecvf.com/
Publication Link : http://www.grvsharma.com/hpresources/mtml_cvpr2016.pdf
Conference Title : CVPR 2016
Publication Title : {LOMo}: Latent Ordinal Model for Facial Analysis in Videos
Topic : Computer Vision
Authors: Karan Sikka and Gaurav Sharma and Marian Bartlett
Conference Website : http://cvpr2016.thecvf.com/
Publication Link : http://www.grvsharma.com/hpresources/lomo_cvpr16_arxiv.pdf
Conference Title : CVPR 2016
Publication Title : Latent Embeddings for Zero-shot Classification
Topic : Computer Vision
Authors: Y. Xian, Z. Akata, G. Sharma, Q. Nguyen, M. Hein, B. Schiele
Conference Website : http://cvpr2016.thecvf.com/
Publication Link : http://www.grvsharma.com/hpresources/arXiv_XASNHS16.pdf
Conference Title : DICTA 2015
Publication Title : Improved Classification and Reconstruction by Introducing Independence and Randomization in Deep Neural Networks
Topic : Computer Vision
Authors: Gaurush Hiranandani, Harish Karnick
Conference Website : http://dictaconference.org/dicta2015/
Publication Link : http://ieeexplore.ieee.org/document/7371270/?reload=true&arnumber=7371270
Conference Title : ICVGIP 2016
Publication Title : Deep Fusion of Visual Signatures for Client-Server Facial Analysis
Topic : Computer Vision
Authors: B. Bhattarai , G. Sharma , F. Jurie
Conference Website : http://www.wikicfp.com/cfp/servlet/event.showcfp?eventid=53456©ownerid=87088
Publication Link : http://www.grvsharma.com/publications.html
Conference Title : ICVGIP 2016
Publication Title : Blind image quality assessment using subspace alignment
Topic : Computer Vision
Authors: Indra Kiran, Tanaya Guha and Gaurav Pandey
Conference Website : http://www.iitg.ernet.in/icvgip2016/
Publication Link :
Conference Title : ICVGIP 2016 | WCVA 2016
Publication Title : A Hybrid Deep Architecture for Face Recognition in Real-Life Scenario
Topic : Computer Vision
Authors: Amartya Sanyal(IIT Kanpur), Dr. Ujjwal Bhattacharya(ISI Kolkata), Dr. Swapan Parui(ISI Kolkata)
Conference Website : https://www.iitg.ernet.in/icvgip2016/WACV/WACVHome.html
Publication Link :
Conference Title : IEEE-ICIP 2016
Publication Title : A trajectory clustering approach to crowd flow segmentation in videos
Topic : Computer Vision
Authors: Rahul Sharma, Tanaya Guha
Conference Website : http://2017.ieeeicip.org/
Publication Link :
Conference Title : IEEE-WACV 2015
Publication Title : Genre and Style Based Painting Classification
Topic : Computer Vision
Authors: Siddharth Agarwal, Harish Karnick, Nirmal Pant, Urvesh Patel
Conference Website : http://wacv2015.org/
Publication Link : http://ieeexplore.ieee.org/abstract/document/7045938/
Conference Title : IEEE/MTS OCEANS
Publication Title : Estimation of Ambient Light and Transmission Map with Common Convolutional Architecture
Topic : Computer Vision
Authors: Young-Sik Shin, Younggun Cho, Gaurav Pandey and Ayoung Kim
Conference Website : http://www.oceans16mtsieeemonterey.org/
Publication Link :
Conference Title : NAACL SRW 2016
Publication Title : Automatic tagging and retrieval of E-Commerce products based on visual Features
Topic : Computer Vision
Authors: Vasu Sharna, Harish Karnick
Conference Website : https://sites.google.com/site/naaclsrw2016/
Publication Link : http://www.aclweb.org/anthology/N/N16/N16-2004.pdf
Conference Title : TPAMI (Journal) 2015
Publication Title : Expanded Parts Model for Semantic Description of Humans in Still Images
Topic : Computer Vision
Authors: G. Sharma, F. Jurie, C. Schmid
Conference Website : http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?reload=true&punumber=34
Publication Link : http://www.grvsharma.com/hpresources/sharma_epm_pami.pdf
Conference Title : WACV 2015
Publication Title : Anomaly Localization in Topic-Based Analysis of Surveillance Videos
Topic : Computer Vision
Authors: Deepak Pathak, Abhijit Sharang, Amitabha Mukerjee
Conference Website : http://wacv2015.org/
Publication Link :
Conference Title : WWW Posters Track, 2016
Publication Title : For the DISTINCT clause of SPARQL queries
Topic : Data Bases
Authors: Medha Atre
Conference Website : http://www2016.ca/
Publication Link :
Conference Title : SIGMOD 2015
Publication Title : Left Bit Right: For SPARQL Join Queries with OPTIONAL Patterns (Left-outer-joins)
Topic : Data Bases
Authors: Medha Atre
Conference Website : http://www.sigmod2015.org/
Publication Link :
Conference Title : CIKM 2016
Publication Title : Quark-X: An Efficient Top-K Processing Framework for RDF Quad Stores
Topic : Data Mining
Authors: Jyoti Leeka, Srikanta Bedathur, Debajyoti Bera, Medha Atre
Conference Website : http://www.cikm2016.org/
Publication Link : http://www.cse.iitk.ac.in/users/atrem/papers/cikm2016.pdf
Conference Title : Coling 2016
Publication Title : Product Classification in E-Commerce using Distributional Semantics
Topic : Natural Language Processing
Authors: Vivek Gupta, Harish Karnick, Ashendra Bansal, Pradhuman Jhala
Conference Website : http://coling2016.anlp.jp/
Publication Link : https://arxiv.org/abs/1606.06083
Conference Title : ICON 2015
Publication Title : Words are not Equal: Graded Weighting Model for building Composite Document Vectors.
Topic : Natural Language Processing
Authors: Pranjal Singh, Amitabh Mukerjee
Conference Website : http://ltrc.iiit.ac.in/icon2015/
Publication Link :
Conference Title : IEEE T-SIPN 2016
Publication Title : Asynchronous Optimization Over Heterogeneous Networks via Consensus ADMM
Topic : Optimization
Authors: Sandeep Kumar (IIT Kanpur), Rahul jain(Qualcomm) and Ketan Rajawat (IIT Kanpur)
Conference Website : http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=6884276
Publication Link : http://ieeexplore.ieee.org/document/7518679/
Conference Title : IJCNN 2016
Publication Title : Adaptive Learning of Dynamic Movement Primitives through Demonstration
Topic : Robotics
Authors: Raj Samant, L. D. Behera and Gaurav Pandey
Conference Website : http://www.ijcnn.org/
Publication Link :
Conference Title : IJRR (Journal) 2015
Publication Title : Learning preferences for manipulation tasks from online coactive feedback
Topic : Robotics
Authors: Ashesh Jain (Cornell University),Shikhar Sharma (IIT Kanpur), Thorsten Joachims (Cornell University), Ashutosh Saxena (Cornell University)
Conference Website : http://ijr.sagepub.com/content/34/10/1296
Publication Link : http://shikharsharma.com/upfiles/jain_sharma_saxena_ijrr2015.pdf
Conference Title : International Journal of Robotics Research
Publication Title : Ford campus vision and lidar data set
Topic : Robotics
Authors: Gaurav Pandey, James R. McBride and Ryan M. Eustice
Conference Website : http://ijr.sagepub.com/
Publication Link : http://robots.engin.umich.edu/publications/gpandey-2011a.pdf
Conference Title : Journal of Field Robotics
Publication Title : Automatic extrinsic calibration of vision and lidar by maximizing mutual information
Topic : Robotics
Authors: Gaurav Pandey, James R. McBride, Silvio Savarese and Ryan M. Eustice
Conference Website : http://www.journalfieldrobotics.org/Home.html
Publication Link : http://robots.engin.umich.edu/publications/gpandey-2015a.pdf
Conference Title : ISRR 2016
Publication Title : Beyond geometric path planning: Learning context-driven trajectory preferences via sub-optimal feedback
Topic : Robotics
Authors: Ashesh Jain (Cornell University) Shikhar Sharma (IIT Kanpur), Ashutosh Saxena (Cornell University)
Conference Website : http://link.springer.com/chapter/10.1007/978-3-319-28872-7_19
Publication Link : http://shikharsharma.com/upfiles/jain_sharma_saxena_isrr2013.pdf
Conference Title : BMVC 2015
Publication Title : Subspace Alignment Based Domain Adaptation for RCNN Detector
Topic : Computer Vision
Authors: A. Raj, V.P. Namboodiri and T. Tuytelaars,
Conference Website : http://bmvc2015.swan.ac.uk/
Publication Link :
Conference Title : BMVC 2015
Publication Title : Adapting RANSAC SVM to Detect Outliers for Robust Classification
Topic : Computer Vision
Authors: S. Debnath, A. Banerjee and V.P. Namboodiri
Conference Website : http://bmvc2015.swan.ac.uk/
Publication Link :
Conference Title : IEEE-FG 2015
Publication Title : Where is my Friend? - Person identification in Social Networks
Topic : Computer Vision
Authors: D. Pathak, Sai Nitish S. and V. P. Namboodiri
Conference Website : http://www.fg2015.org/
Publication Link :
Conference Title : ICASSP
Publication Title : A multimodal mixture-of-expert model for dynamic emotion prediction in movies
Topic : Computer Vision
Authors: Ankit Goyal, Naveen Kumar, Tanaya Guha, and Shrikanth S Narayanan
Conference Website : http://www.icassp2016.org/
Publication Link :
Conference Title : IEEE-ICASSP
Publication Title : A MULTIMODAL MIXTURE-OF-EXPERTS MODEL FOR DYNAMIC EMOTION PREDICTION IN MOVIES
Topic : Computer Vision
Authors: Ankit Goyal (IIT Kanpur), Naveen Kumar(SAIL, USC), Tanaya Guha(IIT Kanpur), Shrikanth S. Narayanan(SAIL, USC)
Conference Website : http://www.icassp2016.org/
Publication Link : ieeexplore.ieee.org/document/7472192/
Conference Title : International Conference on Very Large Data Bases (VLDB Demo)
Publication Title : GARUDA: A System for Large-Scale Mining of Statistically Significant Connected Subgraphs
Topic : Data Mining
Authors: Satyajit Bhadange, Akhil Arora, Arnab Bhattacharya
Conference Website : http://vldb2016.persistent.com/
Publication Link :
Conference Title : CALDAM 2015
Publication Title : Generation of Random Digital Curves using Combinatorial Techniques
Topic : Databases
Authors: Apurba Sarkar, Arindam Biswas, Mousumi Dutt, Arnab Bhattacharya
Conference Website :
Publication Link :
SIGML Reading Group

Reading groups have a thorough discussion on various reseach topics. Often a research paper , book or any relavent research material is discussed. Currently we have two active reading groups i.e. Deep Learning Group and Coginitive Science Group. Below is the details

  • Deep Learning Group
  • Cognitive Science Group
  • `

    Coordinators

  • Name E-mail
    Prabuddha Chakraborty prabudc [at] iitk.ac.in
    Amartya Sanyal asanyal [at] iitk.ac.in
    Piyush Singla piysin [at] iitk.ac.in

    Webmasters

    Name E-mail
    Kshitiz Suman kshitizs [at] iitk.ac.in
    Rishabh Bhardwaj brishabh [at] iitk.ac.in

    Special thank to Previous Coordinators

    Special thank to Previous Webmasters

    Special thank to Previous UX-UI designer

    If you wish to be added to the mailing list you can subscribe by filling the following Form .Outside IIT Kanpur personal can also subscribe subject to approval by filling the Form .If you have any other queries regarding SIGML, please contact at sigml [at] cse.iitk.ac.in