S. Pradeep

About

I am a mathematician examining the structures of dynamical systems, with a fascination for chaotic attractors and the emergence of intelligence. My work merges symbolic systems with geometric deep learning to deepen the theoretical underpinnings of machine learning. I aim to uncover foundational patterns in complex phenomena, advancing computational mathematics and its real-world applications.

Employment

• Indian Institute of Information Technology Surat Assistant Professor (Grade-1)
  2026 - Present
• Indian Institute of Technology Roorkee Postdoctoral Fellow and Principal Investigator
  2022 - 2026
• Indian Institute of Technology Delhi Doctoral Researcher in Symbolic Systems
  2017 - 2022

Education

• Indian Institute of Technology Madras Master of Technology
• IIT Madras Bachelor of Science
• IIT Delhi Master of Science
• IIT Delhi PhD

Projects & Funding

Projects & funding information is unavailable.

Publications (34)

• Foundations of Machine Learning and AI
  2027 DOI: 10.1007/978-3-032-30336-3
  +Save
• Semantic spherical mixup: geometry-aware data augmentation in latent manifolds for parameter-efficient language model adaptation
  Knowledge and Information Systems 2026 DOI: 10.1007/s10115-026-02719-z
  +Save
• Symbolic Ordinal Partition Reservoirs for Model-Free Forecasting of Chaotic Flows
  IEEE Transactions on Emerging Topics in Computational Intelligence 2026 DOI: 10.1109/TETCI.2026.3701390
  +Save
• Python for Mathematical Thinking
  2026 DOI: 10.1007/978-981-95-4080-8
  +Save
• Thalamocortical burst gating enables event-locked memory
  Knowledge-Based Systems 2026 DOI: 10.1016/j.knosys.2025.114947
  +Save
• Hodge-projected echo-state networks with topologically anchored memory for chaotic flows
  Chaos, Solitons & Fractals 2026 DOI: 10.1016/j.chaos.2025.117460
  +Save
• Brain-lateralized reservoirs: Neuro-anatomically constrained echo state networks for chaotic and physiological signals
  Knowledge-Based Systems 2026 DOI: 10.1016/j.knosys.2025.114735
  +Save
• Adaptive group-weighted convolutions: Enhancing equivariant networks via learnable symmetry importance maps
  Neurocomputing 2026 DOI: 10.1016/j.neucom.2026.132689
  +Save
• Phase-modulated permutation reservoirs for robust biomedical time-series forecasting
  Biomedical Signal Processing and Control 2026 DOI: 10.1016/j.bspc.2026.109460
  +Save
• Doubly stochastic inter-assembly coupling via entropic optimal transport in echo-state networks for chaotic flows
  Chaos: An Interdisciplinary Journal of Nonlinear Science 2026 DOI: 10.1063/5.0304827
  +Save