8 papers
Empirical evaluation of the Frank-Wolfe methods for constructing white-box adversarial attacks
Kristina Korotkova, Aleksandr Katrutsa
The construction of adversarial attacks for neural networks appears to be a crucial challenge for their deployment in various services. To estimate the adversarial robustness of a…
Autobidding Arena: unified evaluation of the classical and RL-based autobidding algorithms
Andrey Pudovikov, Alexandra Khirianova, Ekaterina Solodneva +3
Advertisement auctions play a crucial role in revenue generation for e-commerce companies. To make the bidding procedure scalable to thousands of auctions, the automatic bidding (a…
Robust autobidding for noisy conversion prediction models
Andrey Pudovikov, Alexandra Khirianova, Ekaterina Solodneva +4
Managing millions of digital auctions is an essential task for modern advertising auction systems. The main approach to managing digital auctions is an autobidding approach, which…
Dynamic Low-rank Approximation of Full-Matrix Preconditioner for Training Generalized Linear Models
Tatyana Matveeva, Aleksandr Katrutsa, Evgeny Frolov
Adaptive gradient methods like Adagrad and its variants are widespread in large-scale optimization. However, their use of diagonal preconditioning matrices limits the ability to ca…
Cluster Topology-Driven Placement of Experts Reduces Network Traffic in MoE Inference
Danil Sivtsov, Aleksandr Katrutsa, Ivan Oseledets
Efficient deployment of a pre-trained LLM to a cluster with multiple servers is a critical step for providing fast responses to users' queries. The recent success of Mixture-of-Exp…
NNTile: a machine learning framework capable of training extremely large GPT language models on a single node
Aleksandr Mikhalev, Aleksandr Katrutsa, Konstantin Sozykin +1
This study presents an NNTile framework for training large deep neural networks in heterogeneous clusters. The NNTile is based on a StarPU library, which implements task-based para…