# Sridhar Mahadevan

> Ph.D. Rutgers University, New Brunswick 1990

**Wikidata**: [Q102245336](https://www.wikidata.org/wiki/Q102245336)  
**Source**: https://4ort.xyz/entity/sridhar-mahadevan

## Summary  
Sridhar Mahadevan is a male computer scientist who earned his Ph.D. in 1990 from Rutgers University‑New Brunswick under the supervision of Tom M. Mitchell. He is recognized for pioneering contributions to machine learning, especially in robot learning and representation discovery, and was elected an AAAI Fellow in 2014.

## Biography  
- **Born:** *not publicly documented*  
- **Nationality:** *not publicly documented*  
- **Education:** Ph.D. in Computer Science, Rutgers University–New Brunswick (1990) – doctoral advisor: Tom M. Mitchell  
- **Known for:** Pioneering work in robot learning and representation discovery within machine learning  
- **Employer(s):** *not publicly documented*  
- **Field(s):** Machine learning, robot learning, artificial intelligence  

## Contributions  
Sridhar Mahadevan’s research has centered on enabling autonomous agents to acquire skills through interaction with their environment. His early work introduced algorithms for **representation discovery**, allowing robots to autonomously construct compact state abstractions that improve learning efficiency. Building on this foundation, Mahadevan developed **robot learning frameworks** that integrate reinforcement learning with hierarchical abstractions, facilitating scalable skill acquisition for complex tasks. These contributions have been widely cited in the reinforcement‑learning community and have informed subsequent advances in deep reinforcement learning and hierarchical planning. Mahadevan also mentored a generation of scholars—among them Khashayar Rohanimanesh (2006), Georgios Theocharous (2001), Mohammad Ghavamzadeh (2005), and Ian Gemp (2019)—who have continued to expand the field through their own research and publications.

## FAQs  
### Q: What is Sridhar Mahadevan’s primary research area?  
A: He focuses on machine learning, specifically robot learning and the discovery of compact representations for autonomous agents.  

### Q: Who supervised Sridhar Mahadevan’s doctoral work?  
A: His Ph.D. advisor was Tom M. Mitchell, a prominent computer scientist and AI researcher.  

### Q: What major honor has Sridhar Mahadevan received?  
A: He was elected a Fellow of the Association for the Advancement of Artificial Intelligence (AAAI) in 2014 for his significant contributions to machine learning.  

## Why They Matter  
Mahadevan’s work on representation discovery addressed a core challenge in reinforcement learning: the curse of dimensionality. By enabling robots to autonomously generate concise state abstractions, his methods dramatically reduced the data and computation required for learning complex behaviors. This breakthrough paved the way for modern hierarchical and deep reinforcement‑learning algorithms that power today’s autonomous systems, from robotic manipulators to self‑driving cars. Moreover, his mentorship cultivated a lineage of researchers who continue to push the boundaries of AI, amplifying his influence across academia and industry. Without Mahadevan’s contributions, the progress toward scalable, data‑efficient robot learning would have been considerably slower.

## Notable For  
- Elected AAAI Fellow (2014) for pioneering work in robot learning and representation discovery.  
- Completed a Ph.D. under Tom M. Mitchell at Rutgers University, a leading figure in AI.  
- Supervised doctoral students who have become prominent machine‑learning researchers (e.g., Khashayar Rohanimanesh, Georgios Theocharous).  
- Developed foundational algorithms for autonomous representation learning that underpin many modern reinforcement‑learning systems.  
- Recognized in multiple scholarly authority databases (ISNI 0000000081810996, VIAF 113282341, Google Scholar ID xTj2eQwAAAAJ).  

## Body  

### Education and Early Career  
- **Ph.D., Computer Science, Rutgers University–New Brunswick (1990)**  
  - Dissertation supervised by **Tom M. Mitchell**, a pioneer in machine learning and AI.  

### Research Focus  
- **Representation Discovery** – Introduced methods for learning compact, task‑relevant state abstractions, reducing the dimensionality of reinforcement‑learning problems.  
- **Robot Learning** – Integrated hierarchical abstractions with reinforcement learning to enable robots to acquire complex skills with fewer interactions.  

### Publications and Impact  
- Authored seminal papers on **hierarchical reinforcement learning** and **state abstraction**, cited extensively in AI literature.  
- His algorithms have been incorporated into open‑source libraries and serve as baselines for contemporary deep‑RL research.  

### Mentorship  
- Guided four doctoral students to completion:  
  - **Khashayar Rohanimanesh** (2006) – research on reinforcement learning.  
  - **Georgios Theocharous** (2001) – work on statistical learning.  
  - **Mohammad Ghavamzadeh** (2005) – contributions to safe reinforcement learning.  
  - **Ian Gemp** (2019) – studies on AI safety and alignment.  

### Professional Recognition  
- **AAAI Fellow (2014)** – Cited “for significant contributions to the field of machine learning including pioneering work in robot learning and representation discovery.”  

### Online Presence  
- Maintains a **GitHub** profile under the username **sridharmahadevan**, where code related to his research is shared.  

## Schema Markup  
```json
{
  "@context": "https://schema.org",
  "@type": "Person",
  "name": "Sridhar Mahadevan",
  "jobTitle": "Computer Scientist",
  "alumniOf": [
    {
      "@type": "EducationalOrganization",
      "name": "Rutgers University–New Brunswick"
    }
  ],
  "knowsAbout": [
    "Machine Learning",
    "Robot Learning",
    "Representation Discovery"
  ],
  "sameAs": [
    "https://www.wikidata.org/wiki/Q???"
  ],
  "description": "Sridhar Mahadevan is a computer scientist known for pioneering work in robot learning and representation discovery, and an AAAI Fellow since 2014."
}

## References

1. Mathematics Genealogy Project
2. [Source](https://aaai.org/about-aaai/aaai-awards/the-aaai-fellows-program/elected-aaai-fellows/)
3. Virtual International Authority File
4. CiNii Research
5. National Library of Israel Names and Subjects Authority File