I am from Nepal, where I have completed my bachelor’s in computer engineering from Tribhuvan University, Institute of Engineering, Paschimanchal Campus.

Currently, I am working at Palua.Ai Ltd as an AI engineer, where I developed the first ever state-of-the-art Nepali language model trained on a large dataset along with the Nep-gLUE benchmark. I am now working on active learning research for both computer vision and NLP problems to make the data annotation process faster, increase label efficency, and reduce the cost of annotation.

Interests
  • Artificial Intelligence
  • Computer Vision
  • Natural Language Processing
Education

  • Bachelor's In Computer Engineerig, 2017

    Tribhuvan University, Institute of Engineering, Paschimanchal Campus

Publications

Milan Gautam, Sulav Timilsina, Binod Bhattarai (2022). NepBERTa: Nepali Language Model Trained in a Large Corpus. Accepted in (AACL-IJCNLP).

PDF Cite Slides

Projects

*
All Deep Learning Other
Nepali Langauge Model
With the introduction of BERT in 2018, the machine became as capable of understanding natural language as an ordinary human being, making it feasible to sift through massive corpora of text data and extract meaningful information from it.
Milan Gautam
Nepali Speaker Recognition
Speaker Identification is the process of identifying a person based on the audio of their spoken words. Previously, audio processing models were widely accepted, but convolution neural networks have lately demonstrated that they, too, may generate astounding outcomes.
Milan Gautam
Single Image Super Resolution
SISR(Single Image Super-Resolution) is an application of GAN. Image super-resolution is the process of enlarging small photos while maintaining a high level of quality, or of restoring high-resolution images from low-resolution photographs with rich information.
Milan Gautam, Sulav Timilsina
Stock Price forecasting On Nabil Bank
Nabil bank is a bank located in Nepal, it’s been trading in NEPSE(Nepal Stock Exchange ) for the past 20 years. We can easily get this data by going to the NEPSE website.
Milan Gautam
Transactions entity extractor
Because most of today’s world has been digitalized, instead of recording transactions in journals, people record them on their computers. In many circumstances, these transactions contain critical information about the parties involved, thus businesses attempt to obtain it through a variety of means.
Milan Gautam
Facebook Messaging Bot
I used my Facebook conversations to train the RNN model. So the model will aim to mimic how I converse with people, only by providing the initial character for an input, say ‘h,’ will generate an entire message beginning with the character ‘h,’ such as “how are you doing today?
Milan Gautam
Proposal Recommendation and Deployment
DonorsChoose is a United States-based nonprofit organization that allows individuals to donate directly to public school classroom projects. According to last year data around 500,000 proposals were sent by teachers all around the world to DonorsChoose hoping to get the donation for their respective projects.
Milan Gautam
Text Classification using CNN
After learning that an NLP task might be performed using CNN, I was fascinated. So, throughout the study, we will classify emails using CNN. The dataset includes roughly 18000 emails divided into 20 categories such as sports, health, religion, and so on.
Milan Gautam

Recent Posts

Single Image Super Resolution
GAN (Generating Adversarial Network) is about creating, like drawing but completely from scratch. It has got two models: the Generator and the Discriminator are put together into a game of adversary.
Milan Gautam, Sulav Timilsina
Feb 2, 2022 8 min read Deep Learning
Stock Price forecasting On Nabil Bank
In statistical terms, time series forecasting is the process of analyzing the time series data using statistics and modeling to make predictions and informed strategic decisions. It falls under Quantitative Forecasting.
Milan Gautam
Feb 2, 2022 8 min read Deep Learning
Transactions entity extractor
As most of today’s world has been digitalized, instead of writing transactions in journals people do it on their computers. In many cases, these transactions have important information about the parties involved in it, so business tries to acquire it by using many techniques.
Milan Gautam
Jan 3, 2022 9 min read Deep Learning
Facebook Messaging Bot
I have written an medium article examplaning the things that one need to consider when trying to train on your own messages. Please visit here: Code
Milan Gautam
Nov 30, 2020 1 min read Deep Learning

Popular Topics

Deep Learning

Contact