Madan Baduwal

My journey began with a deep fascination for physics, chemistry, and biology-further inspired by philosophy, psychology, cognitive science, and neuroscience,-and a desire to understand the fundamental rules of nature and to solve meaningful problems across domains such as healthcare / medicine, engineering, entertainment, agriculture, social computing, economics & finance….

I am currently pursuing a Ph.D. in Computer Science at Mississippi State University (MSU), where my research spans signal processing, algorithms, machine learning, wireless communications, and robotics. Previously, I served as a Graduate Research Assistant at the University of Texas Permian Basin (UTPB) under the supervision of Prof. Quan Yuan while completing my Master’s degree in Computer Science. I hold a Bachelor’s degree in Computer Engineering from Tribhuvan University, where I worked under the guidance of Prof. Sanjeev Prasad Panday. I have also had the opportunity to collaborate with Mahabir Pun on vision robotics.

English, mathematics, and computer science (CS) soon became the languages through which I explored this curiosity. Today, I use Mathematics (sets, linear algebra, calculus, probability & statistics) and Computer Science (algorithms, machine learning, system design) to explore nature’s principles and develop intelligent systems that solve real-world problems across domains.

I approach problem-solving in three steps(Divide any task into small functional/module):

Input Space

• Mathematics: represent any kind of data / problem as sets / numbers (-infinity … 0 … +infinity)
• Computer Science: store it in data structures like list, dictionary, set, tuple… or finally in bits (e.g., 100101)

Processing

• Mathematics: apply functions (+, -, *, /, etc.)
• Computer Science: apply algorithms to the input space

Output Space

• Mathematics: represent any kind of data as sets / numbers (-infinity… 0 … +infinity)
• Computer Science: store it in data structures like list, dictionary, set, tuple… or finally in bits (e.g., 100101)

Goal-1: Find unified architecture / algorithm of everything(AoE) that captures patterns / representation / Features / Attributes / Descriptors across Observation / Data / Modalities(fields, states > information > Signal: audio, image, video, sensor data… Abstract: text, code, math equations, database entries…), tasks(perception, reasoning, planning, action…), learning(supervised, unsupervised, reinforcement…): over / across time-space / spatio-temporal.

Goal-2: Specialized robots / agents(Mathematical Modeling of Humans / Animals : Cognition Neuroscience, Computer Science, Robotics): Multimodal: Perception / Observation / Understanding, Word Model / State Representation, Memory, Goal / Objective / Reasoning / Planning / Configurator / Path planning / Decision-Making / Policy / Governance and Guardrails, Orchestration and Communication, Tool Interference / Act / Control & Manipulation / Action/Actuation: Feedback / Observation Loop / Reflection / Self-Monitoring, Learning / Continual learning / Active learning / Reinforcement / Adaptation. My ultimate vision is to create specialized robots / agents capable of investigating the fundamental laws of nature and addressing the challenges faced by living beings. In short: Specialized robots / agents loop: Multimodal: Perception → Reasoning → Action → Learning

Goal-3: Prove Goal-1 and Goal-2 (Statements)

If you’re interested in discussing research or potential collaborations, feel free to connect with me on LinkedIn or Twitter, or reach out via email at madanbaduwal100@gmail.com.