Co-located with the ACM India Annual Event, ARCS (formerly IRISS) is a premier ACM India conference spanning two days. The program features:
The audience comprises students, faculty and industry leaders from across India coming together in an interactive environment, such as panel discussions, informal "Ask me Anything" sessions with the ARCS speakers as well as Annual Event invitees like Turing Award laureates.
|2021:||PSG College of Technology|
|2020:||IIT Gandhi Nagar|
|2019:||Rajagiri School of Engineering and Technology - Kochi|
|2017:||Calcutta University/Amity University - Kolkata|
|2016:||Techno Park - Trivandrum|
|2015:||BITS Pilani - Goa|
ACM India announces the 16th Academic Research and Careers for Students
(ARCS) Symposium, Coimbatore, India. Earlier, the event was known as
IRISS. ARCS invites research scholars of computer science and allied areas
in India to showcase their recent work (either published in 2021 or accepted
in 2021 for publication) to a conclave of researchers and potential employers.
Apart from the contributed talks and poster presentations, ARCS 2022 will comprise talks by the ACM-India Doctoral Dissertation Award recipient (DDA), Early Career Research (ECR) Awardee and an invited Keynote Speaker. Further to this there will be four talks by early career researchers discussing the speaker’s transition from PhD to their career and job, their expectations, disappointments, why they chose what they did etc. Finally, we will also have several panel discussions.
Submissions will be in electronic form via EasyChair. Submissions must not exceed 2 pages (including the title page, but excluding bibliography). Submissions should begin with a title followed by the names and affiliations of all co-authors. This should be followed by a detailed abstract and the details of the venue where the work has either appeared or been accepted for publication. Submissions should be no more than two pages long. The usage of pdflatex and the ARCS style file are mandatory; no changes to font size, page geometry, etc. are permitted. Because of ongoing pandemic we will accommodate virtual presentations, if needed. Click << here >> to download the ARCS style file.
Abstract Submission deadline: 10th October, 2021
Notification to Authors: 1 st December, 2021
Download Call for Papers
Sayan Ranu (IIT Delhi), Chair
Ankit Anand (DeepMind)
Amit Awekar (IIT Guwahati)
Nipun Batra (IIT Gandhinagar)
Abhijnan Chakraborty (IIT Delhi)
Ayon Chakraborty (IIT Madras)
Syamantak Das (IIIT Delhi)
Manoj Gupta (IIT Gandhinagar)
Neelima Gupta (University of Delhi)
Aritra Hazra (IIT Kharagpur)
Neeldhara Misra (IIT Gandhinagar)
Adway Mitra (IIT Kharagpur)
Meghna Nasre (IIT Madras)
Swaprava Nath (IIT Kanpur)
Biswabandan Panda (IIT Bombay)
Rohan Paul (IIT Delhi)
Krithika Ramaswamy (IIT Palakkad)
Aishwarya Thiruvengadam (IIT Madras)
Rohit Vaish (TIFR)
Hamim Zafar (IIT Kanpur)
(Chair, IIIT, Bangalore)
Jayant R. Haritsa
(Executive Director, ACM India)
(COO, ACM India)
(PSG TECH, CBE)
(Director & CEO, ACENET Technologies, CBE)
(PSG TECH, CBE)
Dean (Research & Academics), Chitkara University
Peelamedu, Avinashi Road
Coimbatore, Tamilnadu, India
Name: Program Committee
Name: Prof. R. Nadarajan
Abstract: In this talk we will give a brief introduction to the area of Multivariate algorithms. The talk will consist of challenges, excitement, thrills and road ahead of the area, from the perspective of speakers' work.
Abstract: Modern machine learning is characterized by two trends: First, the training data is huge, and often a mixture of several distributions, and model training incurs tremendous energy costs. Second, a model once trained is required to serve diverse real-world settings where the training distribution may not match the test distribution. In this talk, I will discuss current research on handling this distribution mismatch. The talk will span over topics like domain adaptation, domain generalization, out of distribution detection, and robustness.
Abstract: Graphs are all around us, ranging from citation and social networks to Knowledge Graphs (KGs). They are one of the most expressive data structures which have been used to model a variety of problems. Knowledge graphs are structured representations of facts in a graph, where nodes represent entities and edges represent relationships between them. Recent research has resulted in the development of several large KGs; examples include DBpedia, YAGO, NELL, and Freebase. However, all of them tend to be sparse with very few facts per entity. For instance, NELL KG consists of only 1.34 facts per entity. In the first part of the thesis, we propose three solutions to alleviate this problem: (1) KG Canonicalization, i.e., identifying and merging duplicate entities in a KG, (2) Relation Extraction which involves automating the process of extracting semantic relationships between entities from unstructured text, and (3) Link prediction which includes inferring missing facts based on the known facts in a KG. For KG Canonicalization, we propose CESI (Canonicalization using Embeddings and Side Information), a novel approach that performs canonicalization over learned embeddings of Open KGs. The method extends recent advances in KG embedding by incorporating relevant NP and relation phrase side information in a principled manner. For relation extraction, we propose RESIDE, a distantly-supervised neural relation extraction method which utilizes additional side information from KGs for improved relation extraction. Finally, for link prediction, we propose InteractE which extends ConvE, a convolutional neural network-based link prediction method, by increasing the number of feature interactions through three key ideas -- feature permutation, a novel feature reshaping, and circular convolution. Through extensive experiments on multiple datasets, we demonstrate the effectiveness of our proposed methods.
Abstract: In this thesis we consider several (di)graph cut problems and study them from the perspective of parameterized complexity and kernelization. The goal of the study is three-fold: first to extend the otherwise limited understanding of parameterized cut problems on directed graphs; second to extend, and present novel applications of, the existing rich toolkit for undirected cut problems and; third to develop tools that allow the reuse of algorithms to solve the respective problems in the presence of an additional constraint. The concrete questions addressed in the thesis are inspired from some major open problems and concerns in the area. Some of these being the famously active open problem of the existence of a polynomial kernel for Directed Feedback Vertex/Arc Set, sub-exponentiality in FPT beyond tournaments, parameterized algorithms for partitioning problems beyond the classical partitioning problems, the existence of single exponential FPT algorithms for stable versions of classical cut problems and the parameterized complexity of Stable Multicut. We address the above questions either in full, or extend (possibly all) the results known in literature that take steps towards resolving the respective question.
Abstract: This talk is on building compiler infrastructure for emerging programming models for the multicore and accelerator era. The advent of high-productivity programming models and high-performance accelerator chips (especially for machine learning and AI) brings in new challenges on how the next generation of compilers should be built for modularity and reusability. This talk will cover the role of the polyhedral compiler abstraction and techniques towards achieving these goals.