Difference between revisions of "Proposals"

From Research management course
Jump to: navigation, search
(Created page with "List of the proposed projects * '''3D-image reconstruction for the lens-free microscopy''' One has to reconstruct an image of a small biological object (of hundreds micromete...")
 
 
(16 intermediate revisions by one other user not shown)
Line 1: Line 1:
List of the proposed projects  
+
{{#seo:
 +
|title=Proposals
 +
|titlemode=replace
 +
|keywords=Proposals
 +
|description=Proposals: List of the projects  
 +
}}
 +
List of the projects
 +
* '''Brain-computer interface with Functional data analysis''' Human behavioral analysis and forecasting requires models, that have to predict target variables of complex structure. We develop PLS and CCA (Projection to latent space and Canonic correlation analysis) methods towards the Multiview with continuous-time data representation. [[Projects/BCI-FDA: Brain-computer interface with Functional data analysis]]
 +
 
 +
* '''MASSIV: Alternative splicing-inspired protein development''' Eukaryotes have evolved transcription machinery that can augment the protein repertoire without increasing the genome size. The challenge here will be to design an artificial system able to learn the underlying rules of (functional) AS and to generalize to any protein sequence. [[Projects/MASSIV: Alternative splicing-inspired protein development|see details.]]
 +
 
 +
* '''READY-GO: Prediction of protein functional from peptide sequences''' We propose to develop a probabilistic model able to identify such signals and guide the discovery of novel regulatory mechanisms associated with them. The model should be interpretable. [[Projects/READY-GO: Prediction of protein functional from peptide sequences| see details.]]
 +
 
 +
* '''CASF: Comparative assessment of scoring functions for docking problem'''
 +
Pharmacological research is concentrated on constructing a molecule, a ligand, which docks a given protein. This research is expensive since it goes in vitro. We have to propose a deep neural net to forecast the probability of docking, [[Projects/CASF: Comparative assessment of scoring functions|see details.]]
 +
 
 
* '''3D-image reconstruction for the lens-free microscopy'''
 
* '''3D-image reconstruction for the lens-free microscopy'''
One has to reconstruct an image of a small biological object (of hundreds micrometers, like a eucaryotic cell). Its video comes from a device of a brand-new technology: the free lens microscopy. This project combines modelling of interference images and its Fourier transform and GANs, the generative and discriminative deep learning models.
+
One has to reconstruct an image of a small biological object (of hundreds of micrometers, like a eucaryotic cell). Its video comes from a device of a brand-new technology: free lens microscopy. This project combines modeling of interference images and its Fourier transform and GANs, the generative and discriminative deep learning models [[Projects/FLM: Lens-free microscopy image reconstruction|see details.]]
 +
 
 +
* '''Atom-resolution synchrotron image reconstruction'''
 +
For a given synchrotron image one has to reconstruct an object of nano-meter size. There given a [https://en.wikipedia.org/wiki/Reciprocal_lattice reciprocal lattice] image. We have to train a neural net to reconstruct the real and the complex part of a sample object, [[Projects/SRF: Synchrotron radiation facility deep image retrieval|see details.]]
  
* '''CASF: Comparison analysis scoring functions for docking problem'''
+
* '''Long-live health monitoring with wearable devices'''
Pharmacological research is concentrated to constructing a molecule, a ligand, which docks a given protein. This research is expensive since it goes in-vitro. One has to propose a deep neural net to forecast the probability of docking.
+
For daily accelerometer and gyroscope time series, we have to reconstruct typical hand movements and represent these movements as clusters in the phase space.  
  
* '''Atom-resolution synchrotron image reconstruction'''
+
* '''Time series segmentation in low-dimensional space'''
For a given synchrotron image one has to reconstruct an object of nanometer size. There given [https://en.wikipedia.org/wiki/Reciprocal_lattice reciprocal lattice] image. One has to train a neural net to reconstruct the real and the complex part of a sample object.
+
One has to mark up zero-phase segments of various hand movements using electronic watches. Neural nets on spherical harmonics seem to be a good tool.
 +
 
 +
* '''Invariants and compositions of physical activities'''
 +
We have to create a model to describe simple hand movements and compositions of movements. The composition of physical models must fit the composition of machine learning models.
 +
 
 +
* '''Generating WANN: weighted agnostic neural networks'''
 +
We have to generate agnostic nets and ensembles of agnostic nets to solve one of the well-known reinforcement learning problems: controlling toy cars on the road, steps, and pendulums.
 +
 
 +
* '''GAN creates WANN: generative adversarial network generates agnostic networks'''
 +
We have to develop a variational autoencoder, VAE, to generate simple agnostic networks, which are evaluated by some discriminative model.

Latest revision as of 00:33, 17 February 2024

List of the projects

  • Brain-computer interface with Functional data analysis Human behavioral analysis and forecasting requires models, that have to predict target variables of complex structure. We develop PLS and CCA (Projection to latent space and Canonic correlation analysis) methods towards the Multiview with continuous-time data representation. Projects/BCI-FDA: Brain-computer interface with Functional data analysis
  • MASSIV: Alternative splicing-inspired protein development Eukaryotes have evolved transcription machinery that can augment the protein repertoire without increasing the genome size. The challenge here will be to design an artificial system able to learn the underlying rules of (functional) AS and to generalize to any protein sequence. see details.
  • READY-GO: Prediction of protein functional from peptide sequences We propose to develop a probabilistic model able to identify such signals and guide the discovery of novel regulatory mechanisms associated with them. The model should be interpretable. see details.
  • CASF: Comparative assessment of scoring functions for docking problem

Pharmacological research is concentrated on constructing a molecule, a ligand, which docks a given protein. This research is expensive since it goes in vitro. We have to propose a deep neural net to forecast the probability of docking, see details.

  • 3D-image reconstruction for the lens-free microscopy

One has to reconstruct an image of a small biological object (of hundreds of micrometers, like a eucaryotic cell). Its video comes from a device of a brand-new technology: free lens microscopy. This project combines modeling of interference images and its Fourier transform and GANs, the generative and discriminative deep learning models see details.

  • Atom-resolution synchrotron image reconstruction

For a given synchrotron image one has to reconstruct an object of nano-meter size. There given a reciprocal lattice image. We have to train a neural net to reconstruct the real and the complex part of a sample object, see details.

  • Long-live health monitoring with wearable devices

For daily accelerometer and gyroscope time series, we have to reconstruct typical hand movements and represent these movements as clusters in the phase space.

  • Time series segmentation in low-dimensional space

One has to mark up zero-phase segments of various hand movements using electronic watches. Neural nets on spherical harmonics seem to be a good tool.

  • Invariants and compositions of physical activities

We have to create a model to describe simple hand movements and compositions of movements. The composition of physical models must fit the composition of machine learning models.

  • Generating WANN: weighted agnostic neural networks

We have to generate agnostic nets and ensembles of agnostic nets to solve one of the well-known reinforcement learning problems: controlling toy cars on the road, steps, and pendulums.

  • GAN creates WANN: generative adversarial network generates agnostic networks

We have to develop a variational autoencoder, VAE, to generate simple agnostic networks, which are evaluated by some discriminative model.