I am open to collaborate and research on topics relating to Artificial Intelligence, Mathematics, and Computer Science. This page also enlists the publications I have been a part of.

Research Grants

Community Service


Rishit Dagli (2023) Astroformer: More Data might not be all you need for Classification. ICLR 2023 (SOTA on Tiny ImageNet, CIFAR-100) Poster. Video.

Recent advancements in areas such as natural language processing and computer vision rely on intricate and massive models that have been trained using vast amounts of unlabelled or partly labeled data and training or deploying these state-of-the-art methods to resource constraint environments has been a challenge. Galaxy morphologies are crucial to understanding the processes by which galaxies form and evolve. Efficient methods to classify galaxy morphologies are required to extract physical information from modern-day astronomy surveys. In this paper, we introduce methods to learn from less amounts of data. We propose using a hybrid transformer-convolutional architecture drawing much inspiration from the success of CoAtNet and MaxViT. Concretely, we use the transformer-convolutional hybrid with a different stack design for the network and pair it with a careful selection of data augmentation and regularization techniques. Our approach sets a new state-of-the-art on predicting Galaxy Morphologies from images on the Galaxy10 DECals dataset which consists of 17736 labeled images achieving 94.86% top-1 accuracy, beating the current state-of-the-art for this task by 4.62%. Furthermore, this approach also sets a new state-of-the-art on CIFAR-100 and Tiny ImageNet. We also find that models and training methods used for larger datasets would often not work very well in the low-data regime. Our code and models will be released at a later date before the conference.

Rishit Dagli and Ali Mustufa Shaikh (2021) CPPE-5: Medical Personal Protective Equipment Dataset. arXiv Preprint

Won TensorFlow Community Spotlight Award.

We present a new challenging dataset, CPPE - 5 (Medical Personal Protective Equipment), with the goal to allow the study of subordinate categorization of medical personal protective equipments, which is not possible with other popular data sets that focus on broad level categories (such as PASCAL VOC, ImageNet, Microsoft COCO, OpenImages, etc). To make it easy for models trained on this dataset to be used in practical scenarios in complex scenes, our dataset mainly contains images that show complex scenes with several objects in each scene in their natural context. The image collection for this dataset focusing on: obtaining as many non-iconic images as possible and making sure all the images are real-life images unlike other existing datasets in this area. Our dataset includes 5 object categories (coveralls, face shield, gloves, mask, and goggles) and each image is annotated with a set of bounding boxes and positive labels. We present a detailed analysis of the dataset in comparison to other popular broad category datasets as well as datasets focusing on personal protective equipments, we also find that at present there exist no such publicly available datasets. Finally we also analyze performance and compare model complexities on baseline and state-of-the-art models for bounding box results. Our code, data, and trained models are available at https://git.io/cppe5-dataset.

This paper presents a short but non-obvious and interesting theorem in Number Theory that I originally discovered while working on a problem. This theorem states that \( bc - b - c \) is the largest number which \emph{cannot} be written as \( mb + nc \). Given all \( b, c, m and n \in \mathbb{N} \) . In this article I prove the above statement and also show a problem where this theorem could be directly applied to considerably make the problem easier.

Rishit Dagli and Süleyman Eken (2021) Deploying a smart queuing system on edge with Intel OpenVINO toolkit. In: Springer Soft Computing (Previously SOTA on ImagenetVID)

Recent increases in computational power and the development of specialized architecture led to the possibility to perform machine learning, especially inference, on the edge. OpenVINO is a toolkit based on convolutional neural networks that facilitates fast-track development of computer vision algorithms and deep learning neural networks into vision applications, and enables their easy heterogeneous execution across hardware platforms. A smart queue management can be the key to the success of any sector. In this paper, we focus on edge deployments to make the smart queuing system (SQS) accessible by all also providing ability to run it on cheap devices. This gives it the ability to run the queuing system deep learning algorithms on pre-existing computers which a retail store, public transportation facility or a factory may already possess, thus considerably reducing the cost of deployment of such a system. SQS demonstrates how to create a video AI solution on the edge. We validate our results by testing it on multiple edge devices, namely CPU, integrated edge graphic processing unit (iGPU), vision processing unit (VPU) and field-programmable gate arrays (FPGAs). Experimental results show that deploying a SQS on edge is very promising.

Hussain Falih Mahdi and Rishit Dagli and Ali Mustufa and Sameer Nanivadekar (2021) Job Descriptions Keyword Extraction using Attention based Deep Learning Models with BERT. In: IEEE 3rd International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA) (Previously SOTA on QNLI)

In this paper, we focus on creating a keywords extractor especially for a given job description job-related text corpus for better search engine optimization using attention based deep learning techniques. Millions of jobs are posted but most of them end up not being located due to improper SEO and keyword management. We aim to make this as easy to use as possible and allow us to use this for a large number of job descriptions very easily. We also make use of these algorithms to screen or get insights from large number of resumes, summarize and create keywords for a general piece of text or scientific articles. We also investigate the modeling power of BERT (Bidirectional Encoder Representations from Transformers) for the task of keyword extraction from job descriptions. We further validate our results by providing a fully-functional API and testing out the model with real-time job descriptions.

Rishit Dagli (2018) Sierpienski Triangle. In: AMTI 53 rd Annual National Maths Conference, December 2018

This paper derives some new formulas for Sierpienski Fractal and how we can use the derived formulas to make faster computers and efficient cooling chips.The formulas derived can be used in many use cases in field of robotics and electronics. 

Rishit Dagli (2018) Age of Zero. In: AMTI 53 rd Annual National Maths Conference, December 2018

publication description This paper describes the journey of zero and makes an attempt to show who discovered zero and how because the study of past events and inventions can give us many revolutionary ideas.

 Rishit Dagli (2019) Machine Learning as a Decision Aid for Breast Cancer Diagonsis. In: International Advanced Research Journal in Science Engineering and Technology Vol. 6 Issue 10, October 2019

In this paper, we use the diagnosis of breast cytology to demonstrate the applicability of this method to medical diagnosis and decision making. Each of 11 cytological characteristics of breast fine-needle aspirates reported to differ between benign and malignant samples was graded 1 to 10 at the time of sample collection. Nine characteristics were found to differ significantly between benign and malignant samples. Mathematically, these values for each sample were represented by a point in a nine-dimensional space of real variables. We use various different algorithms and also demonstrate the comparison between the algorithms for the classification problem. Finally, an overall accuracy of 99.4048 % is achieved. We only classify 1 % of benign case as malignant. The algorithms used are programmed in python for demonstration purposes. This paper also demonstrates deploying the created model on cloud and building an API for calling the model and verify it.