Ian Goodfellow is known for his innovative work of Generative Adversarial Network and this is an interesting interview on how he has "failed".
As an undergrad at Stanford, Ian struggled in biology and chemistry while preparing for a career in neuroscience.
Google rejected his application for internship when Ian changed his focus to computer science.
Ian wasn't sure he could study deep learning where he was accepted (Stanford and Berkeley) before going to Université de Montréal with Yoshua Bengio as his advisor.
The biggest failure Ian considered was wasting time writing papers which turned out to be dead ends.
Probably the failure I consider the biggest is that I spent most of my PhD trying to solve supervised learning for computer vision using unsupervised feature learning methods, and was caught totally off guard when Alex, Ilya, and Geoff won the ImageNet contest with purely supervised methods
That said, Ian considers his work a success if it influences other researchers even if it gets rejected from a conference. Now researchers have a common way to "influence" by posting their papers on arxiv.org.
Debate experience has helped Ian to overcome such setbacks since "No one is so good that they always win"
Last but not least,
Q: What is the best piece of advice you could give to your past self?
A: I wish I’d used some of those GPUs I bought for deep learning to mine some bitcoin.
Article
Highlights
Ian Goodfellow is known for his innovative work of Generative Adversarial Network and this is an interesting interview on how he has "failed".
As an undergrad at Stanford, Ian struggled in biology and chemistry while preparing for a career in neuroscience.
Google rejected his application for internship when Ian changed his focus to computer science.
Ian wasn't sure he could study deep learning where he was accepted (Stanford and Berkeley) before going to Université de Montréal with Yoshua Bengio as his advisor.
The biggest failure Ian considered was wasting time writing papers which turned out to be dead ends.
That said, Ian considers his work a success if it influences other researchers even if it gets rejected from a conference. Now researchers have a common way to "influence" by posting their papers on arxiv.org.
Debate experience has helped Ian to overcome such setbacks since "No one is so good that they always win"
Last but not least,