Apr 18, 2022
DailyDialog++ and DEB: A new dataset and new metric to make the evaluation of chatbots and dialogue systems better
Ananya B. Sai , Akash Kumar Mohankumar , Siddhartha Arora , Mitesh M. Khapra
How are you feeling today? Good, Great, fine, bad, worst, bored? Either way, you are feeling, you will agree that there is no one right answer to that question and many other millions of questions where multiple answers are correct. While you, as a human, would understand that for some questions there is no single right answer, it’s difficult to teach this to machines! This is one of the problems AI researchers are struggling with today, especially in the chatbot and virtual assistant space.

2 min read

Apr 18, 2022
Generating tailored multimodal fragments from documents and Videos
Dr. Balaji Vasan Srinivasan In today’s digital world, social media and websites have become an important tool for communication. As digital communication is becoming more and more important, companies are looking forward to increase their user engagement to advertise their products, agenda or to just spread knowledge. One of the ways to increase user engagement is to provide them with multimodal content which is a combination of two or more modes of communication such as text, image, video, audio etc.

4 min read

Apr 18, 2022
Scaling graph representation learning algorithms in an implementation agnostic fashion
Prof. Srinivasan Parthasarathy A graph, which consists of nodes and edges, is a pictorial representation of the relationship between various objects. Analysis of graphs can help in inferring social, biological, road, finance and other networks as well. Prof. Srinivasan Parthasarathy who is Professor at Dept. of Computer Science and Engineering and Department of Biomedical Informatics at the Ohio State University and RBCDSAI Distinguished Fellow gave a talk on his research in the area of graph analysis through a talk titled “Scaling graph representation learning algorithms in an implementation agnostic fashion” which was organized as a part of Sixth RBCDSAI Latent View Colloquium on 23rd March 2022.

3 min read

Apr 11, 2022
Controllability of Functional Brain Networks
Prof. Ramkrishna Pasumarthy The brain has undeniably been the most enigmatic organ of the human body. To further our understanding of this mysterious organ, a talk titled “Controllability of Functional Brain Networks” was organized on 2nd September 2021 as a part of the third AI impact seminar. The talk was delivered by Professor Ramkrishna Pasumarthy who is an Associate Professor at IIT Madras. Prof. Pasumarthy started the talk by saying that for a long time most brain modelling was questionnaire-based and only now the availability of huge amounts of FMRI data is driving advances in this field.

4 min read

Mar 28, 2022
Accurate and Interpretable AI models: Towards Deployable AI
Abhishek Ghose , Balaraman Ravindran
A self-driving car bumps into a lamppost. A doctor prescribes the wrong treatment to a patient based on an AI-based diagnostic tool. A missile is misfired by an AI-based defense system. An unfair decision provided by a banking chatbot leads to loss of customers. The above examples are probably sufficient to explain why there is a need to tread carefully while deploying AI-based solutions in the real world. While wrong predictions made by recommendation systems in domains like retail might be inexpensive, such predictions in domains like healthcare, self-driving vehicles, banking or defense can cause hefty monetary losses or even loss of lives.

3 min read

Mar 28, 2022
CAMP: The perfect bacterial matchmaker
Lovers, enemies, friends aren’t the relationships confined to human or animal societies. Tiny beings like bacteria also react to different bacteria in their vicinity, and their growth and survival depend on the kind of relationship they have with the other bacteria residing near them. Two or more types of bacteria living together constitute a microbial community, and in such a community, division of labor occurs which help in the overall growth of community, efficient production of products or teams up to get rid of chemicals which are harmful to the growth of the community.

2 min read

Mar 28, 2022
It takes two to tango — making bio-production more economical
As the world decides to tread on a sustainable path, a more eco-friendly alternative for each and every industrial process is being looked upon. One of the environmentally harmful processes is the production of chemicals through traditional means. Biotechnology offers to remedy this problem by producing chemicals through microbial means, which does not end in creating hazardous byproducts. Despite this advantage, bioproduction hasn’t truly taken off yet due to the poor yield of products obtained through this process.

2 min read

Mar 17, 2022
A new model that solves the charging woes of Electric vehicles
Surendra Reddy Kancharla, Gitakrishnan Ramadurai
As nations race to meet their sustainable goals, the need to shift from fuel-based vehicles to electric vehicles has become more certain. Electric vehicles promise a cleaner and greener future as they don’t release pollutants. Apart from that, electric cars are also more economical to use making them pocket-friendly. Although these advantages make electric vehicles attractive for consumers, yet they are not in demand. This is because they have a short driving range requiring frequent trips to charging stations which is an inconvenience.

3 min read

Mar 9, 2022
Meet Richa Verma
Offline reinforcement learning, a new method of reinforcement learning (RL) is about to make strides in AI technology. The method utilizes the previously collected data for training the algorithm and does not require interaction with the online environment. Richa Verma, a systems engineer at TCS research and PhD scholar at the RBCDSAI-IIT Madras, believes that offline RL has tremendous potential and can scale up systems in a huge way as these offline algorithms can be employed to automate decision-making in various domains like robotics, education and healthcare.

3 min read

Mar 7, 2022
Meet Anusha Kumar
Although science and management are two distinct fields of study, scientific discoveries and inventions have affected business, disrupted markets, and created new trends. Data Science is one such technology that is transforming the way we do business. Machine learning algorithms are helping to predict the stock market trends, credit risks, hire new employees, automate manufacturing plants, etc. This intersection of science and management is what entices Anusha Kumar, a young Data Science researcher who is currently doing her PhD from IIT Madras and working at this interface.

2 min read

Mar 7, 2022
Meet Dr Shreya Goyal
India is a land of many tongues. This linguistic diversity available in the country has attracted linguistic researchers to the country, however, this change in language after every several miles creates a problem of communication for the tourists and travellers. This is where natural language processing, a branch of artificial intelligence, is trying to make a difference. Dr Shreya Goyal, an alumnus of RBCDSAI, IIT Madras and currently a research scientist at American Express, has worked towards solving one such problem in natural language processing for Indian languages through her research.

2 min read

Feb 25, 2022
Learning systems biology, the computational way
The 90s was an interesting time for biology research as researchers shifted from reductionist to a systems approach for studying and understanding biological systems. A decade later, high-throughput technologies revolutionized the field of systems biology by generating heaps of data. Thanks to these technologies, the state of thousands of genes, proteins, chemicals or metabolites at a given time-point, or in response to any stimuli, can be captured now. However, generating insights from this humongous amount of data was like looking for a needle in a haystack.

2 min read

Dec 14, 2021
Meet Prof. Usha Mohan
While sports and academics may seem poles apart to many but Prof. Usha Mohan, who is a Professor at the eminent IIT Madras and a member of RBCDSAI, believes that her involvement in sports is what has provided her with the much-needed resilience required in research. As she teaches and conducts operation research at IIT Madras, she advises her students to dedicate some time in a day for sports as it helps a person to deal with setbacks in a better manner.

3 min read

Oct 28, 2021
Interpretable AI
Prof. Cynthia Rudin For quite a long time, researchers and the public were thrilled by the wonders of machine learning. However, over the period of time, the community realized that the machine learning models aren’t a magic wand and they are as best as the data provided to them during the training and development stage. As the world started making decisions based on AI, there were soon conflicts between human and machine intelligence and therefore the need for explainability of black-box models become apparent.

5 min read

Sep 30, 2021
Machine Learning for Robot Locomotion: Grounded Learning and Adaptive Parameter Learning
Dr.Peter Stone Robot locomotion continues to remain a challenging problem in spite of the advances in the field. To gain more insights on the topic, a talk on “Machine Learning for Robot Locomotion: Grounded Learning and Adaptive Parameter Learning” was organized as a part of the fourth latent colloquium on 30th September 2021. The talk was delivered by Prof. Peter Stone, who is the David Bruton, Jr. Centennial Professor and Associate Chair of Computer Science, as well as Director of Texas Robotics, at the University of Texas at Austin.

3 min read

Sep 20, 2021
SUTRA: Predicting COVID-19 peaks
Prof. M. Vidyasagar, FRS COVID-19 pandemic is testing times for entire humanity and hopes are on science to provide predictions and solutions to wade through and curb this pandemic. Scientists across the world have joined forces and are trying to answer a variety of questions regarding the COVID-19 infection such as how does the infection spread?, how does the virus infect human cells? How can we stop the infection spread?

4 min read

Sep 18, 2021
Mind the messenger: Analyzing the SARS-CoV-2 transcriptome
Coronavirus disease 19 (COVID-19), a severe respiratory disease caused by infection with the novel SARS-CoV-2 virus, has impacted almost every nation globally, making it a pandemic. More than 4.91 billion doses of vaccines have been administered worldwide, but the number of infections continues to rise. mRNA-based vaccines have been developed to bring the pandemic under control in some parts of the world. This brief review will give an overview of how mRNA vaccines work and what we can learn from the transcriptome of COVID-19 patients to identify the significant marker genes responsible for disease severity and mortality.

8 min read

Sep 17, 2021
CombSGPO: A new algorithm to protect wildlife
Aravind Venugopal , Elizabeth Bondi , Harshavardhan Kamarthi , Keval Dholakia , Balaraman Ravindran , Milind Tambe
Poaching and illegal smuggling of wildlife have remained a cause of concern for wildlife authorities. As per the World Wide Fund for Nature (WWF), Wildlife trade poses the second-biggest direct threat to the survival of species after habitat destruction. As per the study conducted by TRAFFIC, a leading wildlife trade monitoring networking of the World Wide Fund for Nature (WWF), around 1,11,312 individual tortoises or freshwater turtles (11,000 a year) have been illegally traded across India since 2009.

3 min read

Sep 16, 2021
Meet Dr. Arun Rajkumar
Biasness and discrimination are known to be human attributes but time and again AI has been in news for its biased behaviour too. Be it in shortlisting candidates for jobs at a reputed firm Amazon or predicting criminals, AI algorithms have failed on the fairness index. While experts understand that the problem lies with the data fed to AI algorithms, the public fears if this technology is used as such it will increase the existing divide in society.

2 min read

Sep 15, 2021
Are DNA variants bugs or features?
Dr. Yana Bromberg With high throughput sequencing technologies available, we can sequence a large number of genomes now and look at the variations in DNA sequences between various organisms and in a population. But how do we identify whether the changes in DNA sequence between two organisms do translate to functional changes or have no effect? This question was taken up for the Eighth IBSE Colloquium talk which was co-organized by RBCDSAI.
DNA , IBSE

2 min read

Sep 15, 2021
Meet Ananya B. Sai
We are surrounded by smart machines! The machines that speak our language, perform mundane tasks for us and solve our queries and problems. Conversational AI is a field that has opened doors for human-machine interaction and soon this technology will enable machines to talk to zillions in their native languages enabling their technology inclusion. But as we move forward to that path, it is of utmost importance to carry out frequent checks to ensure that the chatbots can interpret our query and language correctly and respond accordingly.

2 min read

Sep 13, 2021
A new framework to lend loans in a smarter way
Undoubtedly, loans are one of the major sources of income for banks and other related financial institutions. A loan helps the customers to meet their short term financial needs and the interest given by the customer is the way a bank makes a profit. At once it may seem that extending more loans can increase the profit of such financial institutions, however, laxity in granting loans can lead to more losses as many customers may default on the loan repayment.

2 min read

Sep 13, 2021
Cybercrimes and crime against women increased while other crimes declined during the COVID-19 lockdown
COVID-19 pandemic brought the world to a grinding halt. Nations shut down the non-essential services to cut down the transmission of this new unknown disease. Whilst hospitalizations and mortality were the direct consequences of this undesirable pandemic phase, people suffered due to loss of jobs and for the poor, it became difficult even to make the two ends meet. Experts debated on whether the damage to the economy will be worse than the spread of infection and associated hospitalization and mortality.

2 min read

Sep 3, 2021
The Heads Hypothesis: A Unifying Statistical Approach Towards Understanding Multi-Headed Attention in BERT
Madhura Pande
BERT has been a top contender in the space of NLP models. With its sucess, a parallel stream of research, named BERTology, has emerged, that tries to understand how does BERT work so well. With the similar objective, this blog explains a novel technique for analysing the behaviour of BERT’s attention heads. Read this for interesting insights which unravel the inner working of BERT!

5 min read

Aug 12, 2021
AI Powered Automatic Medical Coding of Electronic Health Records
Sudarsun Santhiappan Jeshuren Chelladurai Medical coding (codifying the medical records) has essentially remained a human powered job since its inception. BUDDI.AI aims to change that by partially automating the process using AI technology. To know more about BUDDI.AI, a talk on “AI Powered Automatic Medical Coding of Electronic Health Records'' was organized as a part of second RBCDSAI AImpact Seminar on 12th August 2021. The talk was delivered by Sudarsun Santhiappan (Co-founder of BUDDI.

2 min read

Jul 21, 2021
Data Commons (The open knowledge Repository)
Dr. Ramanathan Guha A famous mathematician and a data scientist enthusiast, Clive Humby, once said that “Data is the new oil”. Indeed, data is valuable and an untapped resource! However, most of the data currently is in its raw format and can only become valuable once it is cleaned to be used for creating insights. To know how Data Commons is trying to achieve this aim, Dr. Ramanathan Guha, who is a Google Fellow, Vice President at Google and also leading the Data Commons project, was invited to deliver the third RBCDSAI Latent View Colloquium talk on “Data Commons (The open knowledge Repository)” on 21st July 2021.

2 min read

Jul 20, 2021
Meet Dr. Karthik Raman
Fiction as it may seem, a day is just around the corner when we will fix our bodies using the tiny microbes that we may otherwise fear. While we already have probiotics that help us with digestive problems, soon, we will have curds and microbial cocktails to cure anxiety, depression, heart issues and many more. Dr Karthik Raman’s interest lies in these tiny beings! He loves to study these bugs that reside in our body to understand their interactions, complex machinery, and how they function in toto.

4 min read

Jun 22, 2021
Coffee shop banter: Symbolic or Deep Learning? Promising directions of AI
Sparks flew when friends, Prof. Sriraam Natarajan, UT Dallas and Prof. Kristian Kersting, TU Darmstadt decided to get together over a cup of coffee and discuss on the topic “Symbolic or Deep Learning? Promising directions of AI”. The talk started with setting up the stage with “AI is changing the world and has a huge impact in economics, drug development, agriculture etc.” Kristian believes that AI should be an engine for innovation.

3 min read

Apr 10, 2021
Meet Dr. Lelitha Devi
Toxic air and congestion on the road are the grim realities of the metropolitan cities of India. Prof. Lelitha Devi, who has been working in the area of transportation engineering for nearly two decades, believes that a strong push for public transportation backed by technology and policy is required to make it an attractive option for daily commuters. Her team at IIT Madras is working to bring about that change.

3 min read

Mar 14, 2021
Meet Preksha Nema
Curiosity, questioning and understanding sentiments have long been considered human attributes. Preksha Nema, a young scientist’s research is all set to change this! Through her research in natural language processing or popularly called NLP, she is trying to shatter the human-machine boundary by giving computers the magical power to question. Born in Jabalpur city located in the heart of India, Madhya Pradesh, Preksha was fascinated by puzzles and computers from early on.

2 min read

Feb 10, 2021
Levelling up NLP for Indian Languages
Divyanshu Kakwani , Anoop Kunchukuttan , Satish Golla , Gokul N.C , Avik Bhattacharyya , Mitesh M Khapra , Pratyush Kumar
We are working towards building a better ecosystem for Indian languages while also keeping up with the recent advancements in NLP. To this end, we are releasing IndicNLPSuite, which is a collection of various resources and models for Indian languages

6 min read

Feb 9, 2021
Developing India-specific pregnancy dating model from Garbhini cohort
Ashley Xavier , Himanshu Sinha , Nikhita Damaraju
The duration of gestation of pregnancy is the period between the date of conception and date of delivery, which is about 40 weeks of gestation. Estimating gestational age is crucial for accurate prediction of the date of delivery and classifying it as term or preterm, a birth occurring before 37 weeks of gestation. Preterm birth is one of the leading causes of complications for the newborn that can lead to its death.

3 min read

Jan 28, 2021
Meet Dr. Gitakrishnan Ramadurai
Why is everyone heading in the same direction? Is there an accident ahead? When will I reach home? Why can’t cars fly? These are some of the thoughts that cross our minds while stuck in a traffic jam. As time slowly glides, we are nothing but full of frustration. For Dr. Gitakrishnan Ramadurai, however, this is an exciting opportunity! An opportunity to understand the complex, heterogeneous nature of Indian traffic and thinking of the ways artificial intelligence can solve the traffic congestion issues of India.

2 min read

Jan 21, 2021
Fireside chat: with RBCDSAI Distinguished Fellow Prof. Srinivasan Parthasarathy
Ethics in AI is one of the most discussed topics in recent times. Read about the fireside chat with Prof. Srinivasan Parthasarathy where he spoke about AI, Data Science, and Ethics. Get to know the various challenges and solutions embraced by different service providers to keep their clients safe and happy.

3 min read

Jan 20, 2021
Data Science and IoT for addressing ambient air sanctity
Sathish Swaminathan , Raghunathan Rengaswamy
Ambient air quality is a dynamic parameter that shows a high degree of spatiotemporal variation. It is known to impact human life and health. Obtaining meaningful estimates of human exposure to air pollutants is key to mitigating its ill effects. Current air quality monitoring regime consists of expensive, static monitoring stations that are sparsely distributed across a city which in turn results in poor spatial resolution. This particular work, accomplished under project Kaatru, demonstrates the efficacy of low-cost sensor based, vehicle mounted, mobile monitoring paradigm in hyperlocal air quality assessment.

7 min read

Jan 20, 2021
How COVID-19 impacts population movement: A data-driven analysis to study population behavior during a pandemic
Scientific evidence available on the transmission of SARS-CoV-2, the virus that has caused the global outbreak of COVID-19, shows that the disease spreads through droplets launched from an infected person via coughing, sneezing, talking that land on a healthy person in close proximity (less than 6 feet). Epidemiological researchers have found social distancing measures to be very effective in containing the spread of virus in the absence of a proven cure or vaccine.

5 min read

Jan 13, 2021
Ablation-CAM: Making AI trustworthy
Saurabh Desai , Harish G. Ramaswamy
As machine learning is set to change every aspect of our life, a key dilemma plagues the minds of researchers and its users- Can we trust machines to make key life decisions for us? Can we rely solely on the machine to drive us safely to the destination without knowing the basis of its function or shall we allow a machine to operate us instead of a doctor?

4 min read

Jan 13, 2021
Predicting Essential Genes through Network Approach: Deciphering basis of Life
A classic challenge in biology is to study the function of proteins. Of various functions, essential functions are very interesting, as they map to important indispensable genes in an organism. Experimentally identifying these genes is rather expensive and challenging. Computational predictions can help point in the right direction, to prioritise experiments. To date, experimental data are available for <100 organisms! On the other hand, sequencing data are available for 1000s of organisms, as also interactome (networks of interactions) data.

4 min read

Nov 17, 2020
Deep learning in biomedical image analysis
Medical imaging, specifically radiologic imaging is the most commonly used diagnostic tool for disease diagnosis and treatment assessment for a wide variety of conditions. Over the last decades the image acquisition hardware has improved significantly and corresponding image reconstruction software has become more sophisticated. These provide increasingly complex data both in terms of size and content, making it a challenging task for radiologists to sift through and arrive at meaningful diagnosis and therapeutic assessment. The role of AI/ML techniques in this context is to act as a radiologist’s assistant to automate routine tasks and provide preliminary diagnosis. A radiologist can then use the outputs from these systems to speed up and improve accuracy of diagnosis.

3 min read

Sep 5, 2020
Finding Influencers in Social Networks: Reinforcement Learning Shows the Way
Social Network Analysis has given us many tools to effectively manage information dissemination in a social group, study growth and dynamics in such groups, etc. But one of the key challenges when studying social groups of underprivileged or socially marginalized groups is the recovery of the underlying social network itself. This study proposes a machine learning approach for learning to effectively allocate a limited budget to discover the network.

5 min read