Current Position
Assistant Professor of MS&E and, by courtesy, of Computer Science and Electrical Engineering
Intersection with the Project
Overview
The research of Vasilis Syrgkanis forms the pillars of adaptability and analytical intelligence in the agricultural chatbot, particularly regarding user preference learning and understanding the cognitive capabilities of artificial intelligence. The researcher’s work focuses on the intersection between machine learning and causal inference, providing the technical foundation for a system that not only processes information but “learns” dynamically from interaction history and executes complex reasoning necessary for rural property financial management.
”Personalized Adaptation via In-Context Preference Learning” (2024)
This research introduces the Preference Pretrained Transformer (PPT), a framework that allows language model personalization without constant retraining. The central term is Contextual Learning (In-Context Learning - ICL), where the model adapts its behavior by appending recent interaction history to its prompt. Another relevant term is the Contextual Bandit, a scenario where AI makes decisions (such as suggesting a response or action) based on context to maximize utility for the user.
Dynamic Adaptation to Farmer’s Routine
- “PPT leverages the in-context learning capabilities of transformers to dynamically adapt to individual preferences.” “PPT aproveita as capacidades de aprendizado em contexto dos transformers para se adaptar dinamicamente às preferências individuais.”
- Intersection: This technology supports the “Continuous Learning Engine” of the chatbot. Since each farm has specific suppliers and routes, the bot uses WhatsApp conversation history to learn what is important for that farmer without requiring bureaucratic manual configurations.
Continuous Improvement through Interaction
- “The pretrained model interacts with the user, appends the interaction history to its context and generates more personalized responses.” “O modelo interage com o usuário, anexa o histórico de interação ao seu contexto e gera respostas mais personalizadas.”
- Intersection: In the project, users send audio messages about urgent demands. The chatbot uses this technique so that over time, semantic interpretation and alerts become more accurate, reducing up to 80% of failures due to better “knowing” the manager’s habits.
”Discovering Hierarchical Latent Capabilities of Language Models via Causal Representation Learning” (2024)
This work investigates how language model capabilities (such as Claude used in the project) are structured hierarchically. The term Causal Representation Learning refers to identifying hidden factors that explain AI performance. The research demonstrates that complex skills, such as Mathematical Reasoning, depend on fundamental skills like Instruction Following.
Causal Dependency for Financial Management
- “We reveal a clear causal direction starting from general problem-solving capabilities, advancing through instruction-following proficiency, and culminating in mathematical reasoning ability.” “Revelamos uma direção causal clara começando pelas capacidades gerais de resolução de problemas, avançando pela proficiência em seguir instruções e culminando na habilidade de raciocínio matemático.”
- Intersection: For the chatbot to provide cash flow reports with predictive analyses, it first needs to be excellent at following instructions contained in farmer audio. Syrgkanis’ research validates that the success of the bot’s financial automation depends on the AI’s robustness in interpreting simple field commands.
Precision in Operational Command Interpretation
- Mathematical tasks require precise adherence to instructions for correct formatting and problem interpretation, where misunderstandings severely impact accuracy.” “Tarefas matemáticas exigem adesão precisa às instruções para formatação correta e interpretação do problema, onde mal-entendidos impactam severamente a precisão.”
- Intersection: The chatbot extracts tax data from images of receipts and audio. Syrgkanis’ discovery about the causal link between “following instructions” and “mathematical precision” explains why Advanced Audio Processing is the critical functionality of the project: if AI does not correctly interpret the vocal instruction, input material waste and fuel calculations will be compromised.
Full Papers
Vasilis Syrgkanis - Personalized Adaptation via In-Context Preference Learning.pdf Vasilis Syrgkanis - Discovering Hierarchical Latent Capabilities of Language Models via Causal Representation Learning.pdf