My technical/research presentations

Last updated on : 20th October, 2019 

Latest presentations

  1. Presented my ACM WebSci 2019 paper titled “Understanding Brand Consistency from Web Content” at the “Out-of-India” track of India HCI 2019 [Slides]
  2. CNeRG Reading Group talk on 17th October 2019, where I presented the AAAI 2018 paper titled “Weakly Supervised Induction of Affective Events by Optimizing Semantic Consistency”[Slides]

In the previous blog, we discussed some points to remember while preparing for a technical or research presentation. Especially when you have very less amount of time to spare. 

From personal experience in academia

Over the last 2.5 years of my MS degree, both as part of my coursework as well as my research curriculum, I gave some presentations, which I am going to share with you in verbatim. After a great deal of advice and feedback from my seniors and my supervisors, I was able to identify the points of a technical presentation I was blatantly overlooking previously.

You can find the slides in the  Github repo containing some of the presentations I had personally prepared and presented in IIT Kharagpur. Please note that I had not made any modifications or polishing whatsoever after delivering it. So, kindly consider the rough edges :).

The presentations are ordered in terms of oldest to recent.

Presentations made by me

1. An article from the reputed Science magazine :

The spread of true and false news online, published in Science (March 2018 issue). In this presentation prepared by me and Amrith Krishna Da(a Ph.D. scholar, CSE, IIT Kharagpur), we presented the above article. [PPT]

2. My 1st conference paper presentation

My first conference paper was “Understanding Email Interactivity and Predicting User Response to email” and went to present it at Second International Conference on Computational Intelligence, Communications, and Business Analytics (CICBA) 2018 organised at Kalyani Government Engineering College, West Bengal, India.

Here, they already provided a presentation template from beforehand which also included the organisation of the slides.

3. Reading Group (internal) talk at IIT Kharagpur 

Here, I introduce the topic of semi-supervised deep learning techniquesa and present a NIPS 2017 paper in this domain titled “Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results”

4. My own compilation for a Research panel discussion:

Semi-supervised Learning techniques and Active Learning [PPT]
I have only provided my segment, which was a part of a panel discussion covering a broader topic titled Leveraging Unlabeled Data and Environment Access for ML. In the discussion panel, we also covered recent literature in Transfer learning, Zero-shot learning, Reinforcement Learning(with different variants) and finally, Imitation Learning.

The following papers were discussed :

4.1 Semi-supervised learning :

Active Learning for Convolutional Neural Networks, ICLR 2018

Estimating Accuracy from Unlabelled Data, NIPS 2017

When does label propagation fail? a view from a network generative model, IJCAI ‘17

Cost-effective training of deep CNNs with active model adaptation, KDD 2018

4.2 Reinforcement learning

FFNet: Video Fast Forwarding via Reinforcement Learning, CVPR 2018

Human-level control through deep reinforcement learning, Nature 2015

4.3 Transfer, multi-task and few shot learning

One Shot Imitation Learning , NIPS 2017

 When will You Arrive, Estimating Travel Time Based on Deep Neural Networks, AAAI 2018 (Multi-task Learning)

Universal Language Model Fine-Tuning for Text Classification (ULMFit) ACL, 2018

 Deep contextualized word representations (Elmo) NAACL, 2018 ]

A Hierarchical Multi-task Approach for Learning Embeddings from Semantic Tasks AAAI, 2019

 High-risk learning: acquiring new word vectors from tiny data  EMNLP 2017 (short paper)

 Zero-shot Learning of Classifiers from Natural Language Quantification ACL 2018 


5. Reading Group (internal) talk at IIT Kharagpur [Slides]

I and Bidisha Di presented the AAAI 2018 paper titled “Weakly Supervised Induction of Affective Events by Optimizing Semantic Consistency” in the Reading Group of our research group on 17th October, 2019. 

Affective events slide

Please share and provide your feedback in the “Comments” section


Hello everyone. I am Soumyadeep. I have been working on Machine learning projects for the last 4 years. I am now pursuing Ph.D. in Computer Science Department at IIT Kharagpur. I recently completed M.S (Research) from the same department in November, 2019. My research interests involve applying Machine Learning, NLP and Deep Learning to solve Online Reputation Monitoring and Consumer Health Search problems.

Leave a Reply