↓ Skip to main content
Srijan Sood

Publications

378 words·2 mins
Table of Contents

Google Scholar: scholar.google.com · 225 citations · h-index 7

Papers
#

2026
#

  1. TradeFM: A Generative Foundation Model for Trade-flow and Market Microstructure — arXiv S. Sood*, M. Kawawa-Beaudan*, K. Papasotiriou, D. Borrajo, M. Veloso — Under Review

2025
#

  1. FinGraphEx: High Fidelity Financial Knowledge Graph Extraction S. Sood*, K. Papasotiriou*, M. Kawawa-Beaudan, D. Borrajo, M. Veloso — ICAIF 2025

2024
#

  1. AI in Investment Analysis: LLMs for Equity Stock Ratings — arXiv K. Papasotiriou, S. Sood, S. Reynolds, T. Balch — ICAIF 2024

  2. Behavioral Sequence Modeling with Ensemble Learning — arXiv M. Kawawa-Beaudan, S. Sood, S. Palande, G. Mani, T. Balch, M. Veloso — NeurIPS Workshop on Behavioral ML, 2024

  3. Ensemble Methods for Sequence Classification with Hidden Markov Models — arXiv S. Sood*, M. Kawawa-Beaudan*, S. Palande, G. Mani, T. Balch, M. Veloso — arXiv preprint

2023
#

  1. Synthetic Data Applications in Finance — arXiv V. K. Potluru, D. Borrajo, A. Coletta, N. Dalmasso, Y. El-Laham, S. Sood, … — arXiv preprint

  2. Deep Reinforcement Learning for Optimal Portfolio Allocation — PDF S. Sood, K. Papasotiriou, M. Vaiciulis, T. Balch — ICAPS FinPlan 2023

2021
#

  1. Visual Time Series Forecasting: An Image-driven Approach — arXiv S. Sood, Z. Zeng, N. Cohen, T. Balch, M. Veloso — ICAIF 2021

2020
#

  1. Debiasing Classifiers: Is Reality at Variance with Expectation? — arXiv A. Agrawal, F. Pfisterer, B. Bischl, F. Buet-Golfouse, S. Sood, J. Chen, S. Shah, … — arXiv preprint

2018
#

  1. Dungeons and DQNs: Toward Reinforcement Learning Agents that Play Tabletop Roleplaying Games — PDF L. J. Martin, S. Sood, M. O. Riedl — AIIDE 2018

2017
#

  1. Cross-domain Perceptual Reward Functions — arXiv A. D. Edwards, S. Sood, C. L. Isbell Jr — RLDM 2017

* denotes equal contribution

Patents
#

3 granted · 1 allowed · 2 filed

  1. Method and system for forecasting time series by image inpainting — US 12,541,828 (2026)
  2. Method and system for generating keywords — US 12,235,907 (2025)
  3. Method for automatically identifying signals or patterns in time series data — US 11,586,845 (2023)
  4. Method and system for detecting anomalous behavior in stream data — allowed, pending grant
  5. Method and system for fine-tuning LLMs in generating stock ratings predictions — filed (2025)
  6. System and method for classifying behaviors in sequence data — filed (2025)