Masanori Yamada

icon location Tokyo, Japan icon location Ph.D. (science). Researcher of deep learning (robust model, generative model, reinforcement learning).

Results 8 comments of Masanori Yamada

add simple spatial transform attack

OK! My experiment did not replicate the accuracy of the paper in cifar10. If I can check that the values match the author's implementation of tensorflow in cifar10, I will...

add simple spatial transform attack

Just made a PR for a simple spatial transform attack! #82

Is the mutual information estimate is wrong when change the data dimension?

@zhuhaozh Hi, Fig.1 in the paper experiments the effectiveness of Mine in higher dimensions. Mine is better than traditional method in higher dimension but it has a little error. >...

Is the mutual information estimate is wrong when change the data dimension?

@zhuhaozh Sorry for my late reply. I confirmed the same situation. I do not know the cause and investigate(^_^) Sorry, I'm busy, please wait a few weak.

[bug]Cannot override yaml settings on the command line when using the bool option with strtobool.

If you use only the ``--f`` option as in ```cp.add_argument('--f', dest='flag', type=strtointtobool)``` , it works as expected

Add M1 / OSX arm64 wheels

I got the same error with fugashi==1.1.2a6 on my M1 mac. For those having trouble with the same error, I will report the version of fuggashi that worked. fugashi==1.1.1 ipadic==1.0.0...

Failed to render this paper.

https://www.arxiv-vanity.com/papers/2001.11267/ too

PyTorch compatibility

When implemented in PyTorch, linear_sum_assignment in script does not run on the GPU, which causes a speed bottleneck. For example, in STE, linear_sum_assignment exists in the learning loop, which causes...