ServiceNow Research

Hector Palacios

Hector Palacios

Research Scientist

Human Decision Support

Hector Palacios is a Research Scientist at ServiceNow Research. He works on fundamental and applied research on Artificial Intelligence (AI), on the intersection of reasoning and machine learning (ML). At ServiceNow Research, his research has focused on combining pre-trained ML models and reasoning AI techniques like constraint and planning, focusing on generalization and robustness. In academia, Hector worked on automated planning under incomplete information. He holds a PhD in Computer Science from Universitat Pompeu Fabra, Barcelona/Spain, where he was supervised by Hector Geffner. His Ph.D. dissertation received the 2010 Best Dissertation Award by ICAPS and an honorable mention at the 2009 Artificial Intelligence Dissertation Awards by ECCAI, the European Committee for AI. Later on, he shared the IJCAI-JAIR Best Paper Prize 2012 for an “outstanding paper published in JAIR in the preceding five calendar years”.

Interests
  • Planning
  • Dialog
  • Natural Language Processing
  • Reinforcement Learning
  • Compositional Generalization
  • Operations Research
  • Discrete Optimization
  • Multi Step Reasoning
  • Large Language Models
  • Symbolic Reasoning
  • Acting under Uncertainty

Publications

Scaling up ML-based Black-box Planning with Partial STRIPS Models. Workshop at International Join Conference on Artificial Intelligence (IJCAI),  2022.

PDF Cite Code

A Planning based Neural-Symbolic Approach for Embodied Instruction Following. Workshop at the Conference on Computer Vision and Pattern Recognition (CVPR),  2022.

PDF Cite