Intelligent systems

relaxit

[BMM 24-25] "Just Relax It": Implementation of different relaxation methods

Deep-Learning-Course

BMM

BMM

CreationOfIntelligentSystems_Simultaneous_fMRI-EEG

The problem of decoding simultaneously collected EEG-fMRI data is under investigation.

MathematicalForecastingMethods

IDA

Intellectual data analysis course

CreationOfIntelligentSystems_LLM_25

predicator-function-for-neural-networks

swaga

[FDA 25] Spatial-temporal Workflow Automation with Generative Agents

.github

Special profile rep

intsystems.github.io

2025-Project-182

[m1p 2025] Convergence of the loss function surface in transformer neural network architectures

2025_Linich_112

Репозиторий с материалами по НИР. Кафедра Интеллектуальных Систем ФПМИ МФТИ.

bensemble

Учебный проект по байесовскому мультимоделированию студентов первого курса магистратуры 25/26 года обучения. Авторы : Соболевский Федор, Набиев Мухаммадшариф, Василенко Дмитрий, Касюк Вадим

BHPO-AE

selecting an optimal autoencoder architecture using Bayesian optimization

CreationOfIntelligentSystems_KD_25

Firsov_KOT_DARTS

Neural architecture search with target hardware control - continuation of works "Neural architecture search with structure complexity control" - graduate work by student Firsov Sergey and supervisor Bakhteev Oleg

Khabutdinov-MS-Thesis

m1p-MSU-FALL2025

n_ode

Neural ODE classification

nas-for-moe

ODE_with_LLMs

Использование LLM для решения задач, использующих разложение по базису

Papay-MS-Thesis

Rebrikov-BS-Thesis

Rubtsov-BS-thesis

Sign-SGD-and-Differential-Privacy

Code, slides and article on a secure FL algorithm

SToG

SToG is an educational project for the course Bayesian Multimodeling by Eynullayev Altay, Rubtsov Denis, Firsov Sergey and Karpeev Gleb. SToG - is a library for feature selection using different stochastic gating approaches.

text-tree-distance

Repository for research project "Text tree edit distance: a language model-based metric for text hierarchy comparison".

Vector-space-of-generative-models

Задан набор генеративных моделей, описывающий разные выборки/генеральные совокупности данных. Требуется предложить метод векторного представления этих моделей, который будет представлять их статистические свойства.

Zero-shot-structural-pruning

Removing groups of weights from a neural network based on the deep learning computation graph analyzing and estimation of information flow transferred through it.