Department of Computer Science and Engineering

Indian Institute of Technology Hyderabad (IITH) is a premier institute of science and technology established in 2008. IITH has been consistently ranked in the&nbsp;<a href="https://iith.ac.in/reports/" rel="noopener noreferrer">top 10 institutes</a>&nbsp;in India for Engineering&nbsp;<a href="https://www.nirfindia.org/2021/EngineeringRanking.html" rel="noopener noreferrer" target="_blank">according to NIRF</a>&nbsp;making it one of the most coveted schools for science and technology in the country.

IITH offers&nbsp;<a href="https://iith.ac.in/academics/under-graduate/">undergraduate programs</a>&nbsp;in all the classical engineering disciplines, applied sciences, design, as well as several modern interdisciplinary areas. Students are given a flexibility to explore a broad set of areas, and potentially pursue a minor or double major in a discipline that is not their own. Students who wish to seek a deeper understanding of their own discipline are strongly encouraged to get involved in cutting-edge research with the help of a faculty to mentor them, and earn an Honors in their own field.

The very foundation of IIT Hyderabad is based on&nbsp;<a href="https://iith.ac.in/research/">research and innovation</a>. The vibrant research culture is evident from the number of patents and publications that IITH produces consistently every year. IITH offers&nbsp;<a href="https://iith.ac.in/academics/programmes-offered/">graduate programs</a>&nbsp;at both a masters, and a doctoral level, in several diverse areas. There are separate programs for technology, design, science, and liberal arts. The&nbsp;<a href="https://iith.ac.in/academics/post-graduate/">MTech program</a>&nbsp;is offered in various modes and durations to cater to the ever-growing need of postgraduate level professionals in the country.

IITH encourages and supports&nbsp;<a href="https://iic-iith.netlify.app/index.html" rel="noopener noreferrer" target="_blank">innovation and entrepreneurship</a>&nbsp;at every stage. The minor program in entrepreneurship is open to all students of IITH, and many of the courses are offered by industrialists who are entrepreneurs themselves. There are various&nbsp;<a href="https://iith.ac.in/research/centres-incubators/">innovation centers and incubators</a>&nbsp;that promote entrepreneurship and help mentor young innovators. IITH has been very successful in building tie-ups with leading academic institutions around the globe.

IITH enjoys a very special relationship with Japanese universities and industries that goes beyond academic and research collaborations. In fact, some of the iconic buildings in IITH campus will carry the signature of Japanese architecture.

IITH creates a unique holistic ecosystem for education that offers interactive learning, a very flexible academic structure, cutting-edge research, strong industry collaboration, and entrepreneurship. This is an environment which enables students and faculty to translate their dreams into realities.

IIT Hyderabad

Performance of wireless mesh networks (WMNs) in terms of network capacity, end-to-end latency, and network resilience depends upon the prevalent levels of interference. Thus, interference alleviation is a fundamental design concern in multi-radio multi-channel (MRMC) WMNs, and is achieved through a judicious channel assignment (CA) to the radios in a WMN. In our earlier works we have tried to address the problem of estimating the intensity of interference in a wireless network and predicting the performance of CA schemes based on the measure of the interference estimate. We have proposed reliable CA performance prediction approaches which have proven effective in grid WMNs. In this work, we further assess the reliability of these CA performance prediction techniques in a large MRMC WMN which comprises of randomly placed mesh routers. We perform exhaustive simulations on an ns-3 802.11n environment. We obtain conclusive results to demonstrate the efficacy of proposed schemes in random WMNs as well. © 2015 IEEE.

World University root node

2015 International Conference on Computing and Network Communications, CoCoNet 2015

Evaluation of channel assignment performance prediction techniques in random wireless mesh networks

Action recognition in unconstrained videos is one of the most important challenges in computer vision. In this paper, we propose sparsity-inducing dictionaries as an effective representation for action classification in videos. We demonstrate that features obtained from sparsity based representation provide discriminative information useful for classification of action videos into various action classes. We show that the constructed dictionaries are distinct for a large number of action classes resulting in a significant improvement in classification accuracy on the HMDB51 dataset. We further demonstrate the efficacy of dictionaries and sparsity based classification on other large action video datasets like UCF50. © 2016 Elsevier Ltd

Fulltext

Pattern Recognition

Sparsity-inducing dictionaries for effective action classification

In this paper, an approach for human action recognition using genetic algorithms (GA) and deep convolutional neural networks (CNN) is proposed. We demonstrate that initializing the weights of a convolutional neural network (CNN) classifier based on solutions generated by genetic algorithms (GA) minimizes the classification error. A gradient descent algorithm is used to train the CNN classifiers (to find a local minimum) during fitness evaluations of GA chromosomes. The global search capabilities of genetic algorithms and the local search ability of gradient descent algorithm are exploited to find a solution that is closer to global-optimum. We show that combining the evidences of classifiers generated using genetic algorithms helps to improve the performance. We demonstrate the efficacy of the proposed classification system for human action recognition on UCF50 dataset. © 2016 Elsevier Ltd

Human action recognition using genetic algorithms and convolutional neural networks

In this paper, we propose a hybrid deep neural network model for recognizing human actions in videos. A hybrid deep neural network model is designed by the fusion of homogeneous convolutional neural network (CNN) classifiers. The ensemble of classifiers is built by diversifying the input features and varying the initialization of the weights of the neural network. The convolutional neural network classifiers are trained to output a value of one, for the predicted class and a zero, for all the other classes. The outputs of the trained classifiers are considered as confidence value for prediction so that the predicted class will have a confidence value of approximately 1 and the rest of the classes will have a confidence value of approximately 0. The fusion function is computed as the maximum value of the outputs across all classifiers, to pick the correct class label during fusion. The effectiveness of the proposed approach is demonstrated on UCF50 dataset resulting in a high recognition accuracy of 99.68%. © 2015 Elsevier B.V.

Applied Soft Computing Journal

Hybrid deep neural network model for human action recognition

A traditional relational database can evaluate complex queries but requires users to precisely express their information need. But users often do not know what information is available in a database, and hence cannot correctly express their information need. Traditional databases do not provide convenient means for users to gain familiarity with the data. In this paper, we study the problem of exploratory search, which a user may wish to perform to get an understanding of the data set. We note that users often have some decisions already made, so what they need is not an overall database summary, but rather a summary "in context" of the relevant portion of the database. Towards this end, we devise a novel data summarization technique called the Conditional Attribute Dependency (CAD) View, which shows the conditional dependencies between attribute values conditioned on applied selections. The CAD View can help users to gain familiarity with structured datasets in an attribute-wise manner. To evaluate the CAD View, we perform a user study comprising three complex exploratory tasks on a real dataset. Our studies show that users are able to do all the tasks about 4-5 times faster and with better accuracy using the CAD View compared to the data summary shown in faceted navigation, which is currently the most popular search interface for e-commerce and has support for exploratory search. © 2016, Copyright is with the authors.

Advances in Database Technology - EDBT

DBExplorer: Exploratory search in databases

The boxicity (respectively cubicity) of a graph G is the least integer k such that G can be represented as an intersection graph of axis-parallel k-dimensional boxes (respectively k-dimensional unit cubes) and is denoted by box(G) (respectively cub(G)). It was shown by Adiga and Chandran (2010) that for any graph G, cub(G)?box(G)?log;bsubesub?α(G)?, where α(G) is the maximum size of an independent set in G. In this note we show that cub(G)?2?log;bsubesub(G)?box(G)+χ(G)?log;bsubesub?α(G)?, where χ(G) is the chromatic number of G. This result can provide a much better upper bound than that of Adiga and Chandran for graph classes with bounded chromatic number. For example, for bipartite graphs we obtain cub(G)?2(box(G)+?log;bsubesub?α(G)?). Moreover, we show that for every positive integer k, there exist graphs with chromatic number k such that for every ?>0, the value given by our upper bound is at most (1+?) times their cubicity. Thus, our upper bound is almost tight. ©2015 Published by Elsevier B.V.

Discrete Mathematics

Upper bound on cubicity in terms of boxicity for graphs of low chromatic number

Massive Open Online Courses (MOOCs) have initiated a revolution in higher education by providing opportunities for interested students to learn from the comfort of their individual locations at their desired pace. However, an important and highly successful aspect of traditional classroom education/pedagogy, which is modulating content delivery based on understanding real-time student feedback, is conspicuously missing in such e-learning environments. We aim to bridge this gap by proposing a system for automatic recognition of students engagement levels during e-learning sessions, using a crowdsourced discriminative learning approach. The key contributions of this work include a custom dataset for engagement recognition that will be made publicly available with corresponding crowdsourced non-aggregated labels, as well as a novel instance-weighted Multiple Kernel Learning SVM method that can directly consider vote distributions from crowdsourcing platforms in the learning methodology. Our results showed a 14% improvement on the dataset against traditional methods, and a 46% improvement when the most ambiguous class from the dataset is ignored, corroborating the promise of the method. © 2016 IEEE.

2016 IEEE Winter Conference on Applications of Computer Vision, WACV 2016

A crowdsourced approach to student engagement recognition in e-learning environments

The classic K-CYCLE problem asks if a graph G, with vertex set V(G), has a simple cycle containing all vertices of a given set K ⊆ V (G). In terms of colored graphs, it can be rephrased as follows: Given a graph G, a set K ⊆ V (G) and an injective coloring c : K → {1, 2, . . ., |K|}, decide if G has a simple cycle containing each color in {1, 2, . . ., |K|} (once). Another problem widely known since the introduction of color coding is COLORFUL CYCLE. Given a graph G and a coloring c : V (G) → {1, 2, . . ., k} for some k ∈ double-struck N, it asks if G has a simple cycle of length k containing each color in {1, 2, . . ., k} (once). We study a generalization of these problems: Given a graph G, a set K ⊆ V (G), a list-coloring L : K → 2{1, 2, . . ., k∗} for some k∗ ∈ double-struck N and a parameter k ∈ double-struck N, LIST K-CYCLE asks if one can assign a color to each vertex in K so that G would have a simple cycle (of arbitrary length) containing exactly k vertices from K with distinct colors. We design a randomized algorithm for LIST K-CYCLE running in time 2knO(1) on an n-vertex graph, matching the best known running times of algorithms for both K-CYCLE and COLORFUL CYCLE. Moreover, unless the Set Cover Conjecture is false, our algorithm is essentially optimal. We also study a variant of LIST K-CYCLE that generalizes the classic HAMILTONICITY problem, where one specifies the size of a solution. Our results integrate three related algebraic approaches, introduced by Björklund, Husfeldt and Taslaman (SODA'12), Björklund, Kaski and Kowalik (STACS'13), and Björklund (FOCS'10). © Fahad Panolan and Meirav Zehavi.

Leibniz International Proceedings in Informatics, LIPIcs

Parameterized algorithms for list K-Cycle

Feature extraction is one of the most important phases of medical image classification which requires extensive domain knowledge. Convolutional Neural Networks (CNN) have been successfully used for feature extraction in images from different domains involving a lot of classes. In this paper, CNNs are exploited to extract a hierarchical and discriminative representation of X-ray images. This representation is then used for classification of the X-ray images as various parts of the body. Visualization of the feature maps in the hidden layers show that features learnt by the CNN resemble the essential features which help discern the discrimination among different body parts. A comparison on the standard IRMA X-ray image dataset demonstrates that the CNNs easily outperform classifiers with hand-engineered features. © 2016 IEEE.

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

Discriminative feature extraction from X-ray images using deep convolutional neural networks

Let M = (E, I) be a matroid and let S = {S1, ⋯, St} be a family of subsets of E of size p. A subfamily Ŝ ⊆ S is q-representative for S if for every set Y ⊆ E of size at most q, if there is a set X ∈ S disjoint from Y with X ∪ Y ∈ I, then there is a set X ∈ Ŝ disjoint from Y with X ∪ Y ∈ I. By the classic result of Bollob ás, in a uniform matroid, every family of sets of size p has a q-representative family with at most (p p+q) sets. In his famous "two families theorem" from 1977, Lovász proved that the same bound also holds for any matroid representable over a field F. We give an efficient construction of a q-representative family of size at most (p p+q) in time bounded by a polynomial in (p p+q), t, and the time required for field operations. We demonstrate how the efficient construction of representative families can be a powerful tool for designing single-exponential parameterized and exact exponential time algorithms. The applications of our approach include the following: -In the LONG DIRECTED CYCLE problem, the input is a directed n-vertex graph G and the positive integer k. The task is to find a directed cycle of length at least k in G, if such a cycle exists. As a consequence of our 6.75k+o(k)nO(1) time algorithm, we have that a directed cycle of length at least log n, if such a cycle exists, can be found in polynomial time. -In the MINIMUM EQUIVALENT GRAPH (MEG) problem, we are seeking a spanning subdigraph D′ of a given n-vertex digraph D with as few arcs as possible in which the reachability relation is the same as in the original digraph D. -We provide an alternative proof of the recent results for algorithms on graphs of bounded treewidth showing that many "connectivity" problems such as HAMILTONIAN CYCLE or STEINER TREE can be solved in time 2O(t)n on n-vertex graphs of treewidth at most t. For the special case of uniform matroids on nelements, we give a faster algorithm to compute a representative family. We use this algorithm to provide the fastest known deterministic parameterized algorithms for k- PATH, k-TREE, and, more generally, k-SUBGRAPH ISOMORPHISM, where the k-vertex pattern graph is of constant treewidth. © 2016 ACM.