Gautam Singh

lastname.firstname@rutgers.edu · CV · Twitter · LinkedIn · Github · Scholar

 

I am a Ph.D. candidate in the Department of Computer Science at Rutgers University, advised by Prof. Sungjin Ahn.

I am interested in perception and generative modeling in unstructured modalities (e.g., vision), focusing on compositionality and out-of-distribution (OOD) generalization. I pursue unsupervised learning approaches for harnessing the full potential of all available data and large-scale training. Toward perception, my focus is on how to extract reusable and composable primitives such as objects and attributes from raw observations. Toward generative modeling, I am interested in building compositional world models for applications in agent learning.

I spent the summer of 2021 as an Applied Scientist Intern at Amazon. Before starting a Ph.D. in 2018, I worked at IBM Research India. In 2015, I received a Bachelor of Technology degree from IIT Guwahati. I also spent a summer at KAIST as a research intern in 2014.

 

---

Recent News

• Sep 2023: Joined as a research intern at NVIDIA Research Autonomous Vehicle Research Group collaborating with Gerry Che and Yue Wang.
• May 2023: Gave a talk at the Toronto AI in Robotics (AIR) seminar, focusing on representation learning for systematic generalization. YouTube link here!
• Feb 2023: We have released the code and datasets for our recent work Neural Systematic Binder here!
• Jan 2023: Our recent work, Neural Systematic Binder, has been accepted at ICLR 2023.

---

Selected Publications

Equal contributions are denoted using {}.

2023

• Imagine the Unseen World: A Systematic Visual Imagination Benchmark
  
  • {Yeongbin Kim, Gautam Singh}, Junyeong Park, Caglar Gulcehre, Sungjin Ahn
  • NeurIPS 2023 

• Object-Centric Slot Diffusion
  
  • Jindong Jiang, Fei Deng, Gautam Singh, Sungjin Ahn
  • NeurIPS 2023 (Spotlight) [pdf]

• Neural Systematic Binder
  
  • Gautam Singh, Yeongbin Kim, Sungjin Ahn
  • ICLR 2023 [pdf] [project] [code] [data]

2022

• Simple Unsupervised Object-Centric Learning for Complex and Naturalistic Videos
  
  • Gautam Singh, Yi-Fu Wu, Sungjin Ahn
  • NeurIPS 2022 [pdf] [poster] [project] [code] [data]

• Illiterate DALL-E Learns to Compose
  • Gautam Singh, Fei Deng, Sungjin Ahn
  • ICLR 2022 [pdf] [project] [code]

2021

• Structured World Belief for Reinforcement Learning in POMDP
  • Gautam Singh, Skand Peri, Junghyun Kim, Hyunseok Kim, Sungjin Ahn
  • ICML 2021 [pdf] [project]

2020

• Robustifying Sequential Neural Processes
  • Jaesik Yoon, Gautam Singh, Sungjin Ahn
  • ICML 2020 [pdf]

• SPACE: Unsupervised Object-Oriented Scene Representation via Spatial Attention and Decomposition
  • {Zhixuan Lin, Yi-Fu Wu, Skand Vishwanath Peri}, Weihao Sun, Gautam Singh, Fei Deng, Jindong Jiang, Sungjin Ahn
  • ICLR 2020 [pdf] [project]

2019

• Sequential Neural Processes
  • {Gautam Singh, Jaesik Yoon}, Sungjin Ahn
  • NeurIPS 2019 (Spotlight) [pdf] [project] [code]

---

Media

• Oct 2021: Rutgers University’s AI Researchers Propose A Slot-Based Autoencoder Architecture, Called SLot Attention TransformEr (SLATE)

---

Service

• Conference Reviewer: NeurIPS23, ICLR23, NeurIPS22, ICML22
• Workshop Reviewer: OSC@ICLR22