Dmitrij TichonovCUDA Series: Streams and SynchronizationEfficient management of concurrent tasks is essential for maximizing the performance of GPU-based applications. Streams allow to execute…4d ago4d ago
Dmitrij TichonovCUDA Series: Memory and AllocationEfficient memory management is critical for maximizing performance. CUDA provides various mechanisms for allocating memory on both the host…4d ago4d ago
Dmitrij TichonovCUDA Series: Execution and GraphsIn GPU programming, managing complex workflows with multiple dependent kernels and memory operations can lead to significant overhead. CUDA…4d ago4d ago
Dmitrij TichonovAI in Action: Designing a Neural Network for Bot DetectionLets set the stageSep 151Sep 151
Dmitrij TichonovMachine learning at scaleAt a very high level when you are training a deep learning model with an optimisation algorithm such as gradient descent, usually process…Nov 1, 2017Nov 1, 2017
Dmitrij TichonovDebunking loss functions in deep learningAt the heart of any supervised machine learning algorithm there is a training phase which could not be done without an optimisation…Oct 24, 20171Oct 24, 20171