Arslan Chaudhry
email: arslan.chaudhry [AT] eng.ox.ac.uk

| CV | Google Scholar | Rhodes |
| Github | LinkedIn | Twitter |

I will be joining Google DeepMind as a Research Scientist!

I recently finished my PhD in Machine Learning with Prof. Philip H. S. Torr at the University of Oxford. I was funded by the Rhodes Trust. Before joining Oxford, I worked as a software developer at Mentor Graphics for more than three years developing Mentor Embedded Hypervisor. Prior to that, I did my undergraduate in Electrical Engineering from the University of Engineering and Technology, Lahore where I graduated at the top of my class.

My research focuses on studying Continual Learning for efficient Machine Learning. I interned with Facebook AI and Google Research during my PhD.

 
  News
 

  Publications
  1. Using Hindsight to Anchor Past Knowledge in Continual Learning
    Arslan Chaudhry, David Lopez-Paz, Albert Gordo, Puneet K. Dokania, Philip H. S. Torr
    AAAI, 2021
    [Paper]
  2.  
  3. Continual Learning in Low-rank Orthogonal Subspaces
    Arslan Chaudhry, Naeemullah Khan, Puneet K. Dokania, Philip Torr
    NeurIPS, 2020
    [Paper] [Code]
  4.  
  5. On Tiny Episodic Memories in Continual Learning
    Arslan Chaudhry, Marcus Rohrbach, Mohamed Elhoseiny, Thalaiyasingam Ajanthan, Puneet K. Dokania, Philip H. S. Torr, Marc'Aurelio Ranzato
    MTLRL Workshop (ICML), 2019
    [Paper] [Code]
  6.  
  7. Efficient Lifelong Learning with A-GEM
    Arslan Chaudhry, Marc'aurelio Ranzato, Marcus Rohrbach, Mohamed Elhoseiny
    ICLR, 2019
    [Paper] [Code]
  8.  
  9. Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigence
    Arslan Chaudhry, Puneet Kumar Dokania, Thalaiyasingam Ajanthan, Philip H.S Torr
    ECCV, 2018
    [Paper] [Code]
  10.  
  11. Discovering Class-Specific Pixels for Weakly-Supervised Semantic Segmentation
    Arslan Chaudhry, Puneet Kumar Dokania, Philip H.S Torr
    BMVC, 2017 (oral)
    [Paper] [Code] [Talk]

[Workshops]

  1. Continual Learning with Tiny Episodic Memories
    Arslan Chaudhry, Marcus Rohrbach, Mohamed Elhoseiny, Thalaiyasingam Ajanthan, Puneet K. Dokania, Philip H. S. Torr, Marc'Aurelio Ranzato
    ICML, 2019 (Workshop on Multi-Task and Lifelong Reinforcement Learning)
    [Paper] [Code]
  2.  
  3. On Efficient Lifelong Learning with A-GEM
    Arslan Chaudhry, Marc'Aurelio Ranzato, Marcus Rohrbach, Mohamed Elhoseiny
    NeurIPS, 2018 (Continual Learning Workshop)
    [Paper] [Code]
 

  Experience
 

  Teaching
  • Tutorials, Machine Learning, Trinity 2018, Hilary 2020, Stanford House
  • Tutorials, Networking, Trinity (2018, 2019), University of Oxford
  • Tutorials, Operating Systems, Hilary (2018, 2019), University of Oxford
  • Lab, Software Engineering, Hilary (2017, 2018, 2019), University of Oxford
  • Teaching Assistant, Operating Systems, Spring 2013, UET
 

  Talks
  • Continual Learning in Deep Neural Networks, July 2020, KAUST
  • Continual Learning in Deep Neural Networks, July 2020, Google Research
  • Continual Learning in Deep Neural Networks, April 2020, Google DeepMind
  • Machine Learning Introduction, December 2017, UET Lahore
 

  Academic Service
  • Conference Reviewer: ICML (2019-2020), NeurIPS (2019-2020), ICLR (2021), CVPR (2019-2020), ICCV (2019), ECCV (2020)
  • Journal Reviewer: IEEE TPAMI, IEEE Transaction on Multimedia
 

Template: this, this, this and this