Madan Baduwal

200 E

Interstate-20

Odessa, Texas 79766

I am a final-year Computer Science(M.S) Student at The University Of Texas Permian Basin(UTPB), conducting research under Prof. Quan Yuan in machine learning and deep learning, particularly in transformer-based architectures and diffusion models. My research interests lie in Cognitive neuroscience, Data Structure and Algorithms, Deep Learning(Multimodal Learning, Generative AI, LLM), Computer Vision, Natural Language Processing,Software Engineering, and Robotics. In particular I am interested in reverse-engineering the human mind to build embodied machines that can see, listen, speak, learn, and interact with human-level cognition and social intelligence.

I earned a BE in Computer Engineering at Tribhuvan University , under the direction of Sanjeev Prasad Panday.I am lucky to have opportunities to work with Mahabir Pun , Rabin Dhakal. If you are interested in discussing research, feel free to reach out via LinkedIn, or Twitter. You can also send me an email: madanbaduwal100@gmail.com.

news

Jan 16, 2024	Started working as a Graduate Assistant at UTPB. 💁
Mar 15, 2023	Began working as an Orientation Leader at UTPB.
Mar 10, 2023	Joined as an Undergraduate Student Assistant Editor at UTPB.
Jan 11, 2023	Started Master of Science in Computer Science at the UTPB. 🎓
Aug 17, 2022	Started Life Long Learning.📖

latest posts

Dec 22, 2020	Covid-19 Detection
Dec 10, 2020	Python functions

selected publications

SigV

Signature Verification using Convolutional Neural Network and Autoencoder.

Prakash Ratna Prajapati , Samiksha Poudel , Madan Baduwal , and 2 more authors

Journal of the Institute of Engineering, 2021

Abs PDF

Signature has been one of the widely used verification biometrics out there. Handwritten signatures are used in cheques, forms, letters, applications, minutes, etc. The Signature of every individual is unique in nature, that is why it is essential that a person’s handwritten signature be uniquely identified. Signature Verification is a widely used method for authenticating any individual during absence. Human verification is prone to inaccuracy and sometimes indecisiveness. This paper presents an investigation of using Convolutional Neural Network (CNN) for Writer-Dependent models in signature verification. Random distortions were generated in genuine images using an autoencoder to get forged signatures, which were passed to the classifier during training. The paper details all the pre-processing steps carried out on the image and shows various test results for changing the number of training sets of images. The average test accuracy for Persian dataset is 83% when the system was trained with 22 genuine images. There was a decrease of 9.4% in accuracy when the model was trained with 9 genuine images.