Sneaky Machine Learning Reading Group

Presenter: Jialin

Date: 31st Jan

Minutes:

He: original GAN can hardly learn the mixture of 2 distributions well.

Jialin: U-net concatenates left to the right at the same layer.

Tom: these connections maintain spatial resolution.

He: why weight of two ways is not a learnable parameter?

Jialin: Always fix the earlier layers when train the new layers.

He: Now that use means why calculate stand deviation?

Tom: Low resolution determines features.

Presenter: Zheyan

Date 7th Feb

Contents:

1 How to do convolution for dimension 0-3 tensors.

2 The overall convolution filter for a rank n tensor is a rank 2n tensor and why decompose it.

3 Some methods to decompose filters for n dimensions tensors.

Presenter: Dody

Date 15th Feb

After the presentation please try to give your opinion on the paper!Contents:

TempoGAN projects the low resolution simulation result to high resolution result. Compared with past CNN, this module take 3 frames in time series as input.

Discussion:

Why there is dimension inconsistency?

Some region of TempoGAN result is worse than baseline. It is hard to say which result is better.

It is better to present one paper with more details than 2 papers.

Presenter: Feixiang

Discussion:

The words are embedded into ID format.

What is the training time?

The encoder and classifier are trained individually.

How to separate speech information and left words information.

Is there evidence  that shows disentangle really works?

In the video there is minor change other than the mouse and bleezing eyes.

Presenter: Baiyi

Discussion:

This is CycleGAN, networks used for creating dataset.

Constraint GAN means difficulty in training.

Possible build two datasets, with and without makeup.

The purpose of attack: avoid recognition, update defense system.

Presenter: Deshan

Main content:

The difference of knit and woven and the reason to do yarn level simulation.

What are the Lagrangian and Euler DoFs.

The contact plane is used to compute contact force.

What is visco-elastic property.

After presentation everyone please try to give some opinion on the paper.

Presenter: Yunfeng

Presenter: Xue

Paper: GAN Compression -Efficient Architectures for interactive Conditional GANs.

Presenter: Jialin

Contents:

Traditional GAN conditional GAN and Pix2Pix GAN.

Previous methods to reduce calculation

Validation to choose proper student NN.

Decouple training and search.

Presentor: Zheyan

Contents:

Flow control and active control background, it is hard to design active control.

experiment setting.

Reinforcement algorithms. The output of ANN is the distribution of control strategy. Optimization loss is the long term performance. The samplle strategy number increase over time and seems training expense continues to grow.

Presenter : Dody

Contents:

Can we use ML to accelerate hypeelastic simulation?

Traditional Reduce order method use PCA.

How to integrate autoencoder.

Questions:

Why do not use autoencoder alone but with PCA?

What does J means?

Presentor: Feixiang

Presenter: Deshan

Contents:

Concepts of discretization, implicit Euler , collision detection and responce

It usually need labor work to adjust the simulation parameters. The work aims automatically adjust parameters.

Dynamic and Continuous collision dectection, QR decomposion

Derivatives of the physics solve. Constrains, KKT condition.

Experiments

Presenter: Jialin

Contents:

After the presentation please try to give your opinion on the paper!Traditional image blending.

Combine MVS and IBR.

Deep Blending, 4 challenges in old methods.

Two offline pipeline, one online.

3D geometry reconstruction.

DNN architecture.

Rendering algorithm.

Two novel training loss.

Limitation.

After the presentation please try to give your opinion on the paper!

Presenter: Zheyan

MeshingNet by Zheyan

The question asked in the presentation:

1. How is the algorithm of the solution?
2. How does the network connect to the mesh?
3. When it breaks the triangle?
4. Does it resize the triangle?
5. Is the magic generator differentiable/ inside or outside the neural network?
6. How do you make the particular neural network design?
7. Why do you show training loss, rather than testing loss?
8. If you don’t show the testing with training, it’s hard to tell.

The comments on every slide:

2^nd slide: To make the figures with less information/To make it a bit simpler.

3^rd slides: What data are you actually training on? Use a real mesh instead of simplified mesh.

4^th slide: what does 3000 mean?

6^th slide: It needs motivation: how should it affect the mesh? Need examples.

7^th slide: what’s the baseline?

10^th slide: make it look 3d.

11^th slide: Why do the result by CZ have colour?

by Yunfeng

Paper link:

After the presentation please try to give your opinion on the paper!

Presenter: Yunfeng

Minutes:

• Deshang : "why the algorithm 1 returns right value of binary search , instead of others, like middle value or left value?"
  • Yunfeng: I think it should return Vright or Vleft(it depends on if f(x0 + vmid theta)) == y0. And for binary search, the Vleft/Vright has been updated by Vmid already, so we don't need to return Vmid
• He Wang:  "It is very interesting to see that they can prove an upper bound of the number of iterations needed. It would be good to show the proof if possible."
• Jialin :  "They do not consider time series data. I agree that we need to consider some particular problems like weighting scheme when we attack time-series data."
  • The last paragraph in 3.2 mentioned that for high-dimensional problems, they sample 20 vectors from Gaussian distribution and average their estimators to get gˆ. This seems not to give any theoretical basis and seems to be based entirely on their experimental results? 
• Feixiang:  "we can use a vector to indicate the direction, theta, but how to decide the length of the vector, is it with the same dimension of the input image?"
• Dody : "they said: for high-dimensional problems, we found the estimation in equation (7) is very noisy" -> would u explain, what do they mean?
  • Yunfeng : they do the estimation of g-hat  q times,then average them. 

Presenter: Dody Dharma

Paper link:

Presenter: Feixiang

This research proposes a vision-based deep neural network for estimating cloth/substrate dry friction coefficient. It uses synthetic datasets to train the neural network model which, however, is capable of estimating dry friction coefficient of real cloth. This is realised by accurately recover the cloth dynamics related to the dry friction coefficient. Moreover, this research introduces Conditional Friction Model which is optimised for extracting the information related with dry friction coefficient.

Jialin presents:

The back ground is to obtain HD images from low resolution. Pulse ensures the quality of the super resolution image while tradition methods lead to blurring, they only compare HR and LR images. Traditional MSE neglect detail such as texture so that they are smoothed out. To get rid it people increase distance between SR and HR or use perceptual loss.

This paper has new paradigm and use novel method including Gaussian domain search.

PULSE may illuminate some biases inherent in StyleGAN.

CVPR video/paper for discussion:

Zheyan present:

CVPR video for discussion:

Dody will present:

CVPR video/paper for discussion:

Deshan presents:

Looking this video:

Feixiang presents:

Looking this video:

Jialin presents:

Zheyan presents:

Deshan presents:

Dody talked on graph_CNN and how this technique may help our research.

Zheyan Zhang presents.

By Deshan

Presenter: Dody

Presenter: Feixiang

Ricardo Luna Gutierrez

presents

Kyle present

present:

By Jiangbei

By Sara

By Deshan

Recent works on human motion classification attack

By He

By Dody

by Feixiang

By Maria

Markus presents:

Jialin present:

Shaun present:

the field of my research with an overview of the problem and the current state of research surrounding it

He Wang present:

Tom present:

Baiyi present:

Maria present:

Zheyan present:

He present:

Usman present:

Baiyi present:

zheyan present:

Finite element with machine learning

Maria present:

Jiangbei present:

Baiyi present:

Maria present:

Shaun present:

Mou Li present:

Usman present:

Feixiang present:

Feixiang present: