EINNET - Optimizing Tensor Programs with Derivation-Based Transformations
Optimizing tensor programs for deep neural networks using derivation-based transformations, outperforming existing optimizers by leveraging a larger search space and automatically creating new operators.
Effectively Scheduling Computational Graphs of Deep Neural Networks toward Their Domain-Specific Accelerators
Innovative approach for scheduling DNN computational graphs on domain-specific accelerators, optimizing resource utilization and enhancing performance through synergy between network and hardware architectures.
Decentralized Application-Level Adaptive Scheduling for Multi-Instance DNNs on Open Mobile Devices
Explores decentralized scheduling for multiple DNN-powered apps on open mobile systems, using Deep Reinforcement Learning to achieve balanced performance gains and energy savings across various scenarios.
Boosting Research with Machine Learning
Explore machine learning applications in research, from neuron recognition to EEG-based movement prediction. Learn Python techniques for data preparation and analysis using various ML methods.
Computer Vision Using Go and OpenCV 3
Explore computer vision applications using GoCV, combining Go and OpenCV. Learn to build advanced IoT solutions with face tracking, motion detection, and drone control using Tensorflow.
Deep Learning in the World of Little Ponies
Explore computer vision and deep neural networks to build custom image classifiers, with a fun twist using "My Little Pony" characters as examples.
The Topology of Deep Neural Networks - Designing Your Model
Explore deep neural network topology and model design techniques for effective machine learning implementations.
How Computers Are Learning to Be Creative
Explore how neural networks can generate art and poetry, blurring the line between perception and creativity in machine learning. Witness captivating demos of AI-generated visual and textual content.
Input-Relational Verification of Deep Neural Networks
Explore novel input-relational verification techniques for deep neural networks using difference tracking and the DiffPoly abstract domain. Gain insights into improving precision for various properties.
Architecture-Preserving Provable Repair of Deep Neural Networks
Explore architecture-preserving V-polytope provable repair of deep neural networks. Learn about a novel approach that guarantees specification satisfaction while preserving DNN architecture, supporting various layers and activation functions.
MAGPY: Compiling Eager Mode DNN Programs by Monitoring Execution States
Optimizing deep learning programs by generating complete operator graphs through runtime monitoring, enhancing performance for complex Python-based DNN implementations.
Centimani - Enabling Fast AI Accelerator Selection for DNN Training
Innovative performance predictor for selecting optimal AI accelerators in DNN training, enhancing efficiency and decision-making in heterogeneous hardware environments.
DeepTheft: Stealing DNN Model Architectures through Power Side Channel - 2024
Explore DeepTheft, a novel technique for extracting DNN model architectures through power side channel analysis, presented by Garrison Gao at IEEE.
Multi-Precision Optimization Algorithm: Decreasing Computational Cost and Controlling Computational Error
Explore multi-precision optimization algorithms to reduce computational costs and control errors in smooth, non-convex function minimization, with applications in deep neural network training.
DP-GEN: An Optimal Active Learning Protocol for Constructing Interatomic Potentials - Materials Simulation
Explore DP-GEN, an innovative active learning protocol revolutionizing molecular modeling by constructing accurate interatomic potentials. Learn its three-step iterative process and applications in materials simulation.
Unveiling Hidden Backdoors in Manifold Distribution Gaps
Explore backdoor attacks in deep neural networks, focusing on manifold distribution gaps. Learn to generate stealthy triggered samples and extend decision boundaries for enhanced security understanding.
Pangu-Weather: Accurate Medium-Range Global Weather Forecasting with 3D Deep Neural Networks
Explore Pangu-Weather, an AI-based method for accurate medium-range global weather forecasting using 3D deep networks and Earth-specific priors, outperforming traditional numerical prediction systems.
Modeling Individuals Without Data via Secondary Task Transfer Learning
Discover a novel transfer learning method for modeling individual behavior using limited data from secondary tasks, outperforming standard approaches in financial health and game design.
Financial Client Churn Detection Using Computer Vision
Detect financial client churn using computer vision techniques. Convert historical client data into images and apply deep neural networks for predictive analysis.
Machine Learning for Trust and Safety at Pinterest
Explore Pinterest's ML solutions for Trust and Safety, including DNN, CNN, and GCN techniques. Learn about challenges, design principles, and processes for developing and deploying effective content moderation systems.
I-Vector Representation Based on GMM and DNN for Audio Classification - 2015
Explore I-vector approach for audio classification, including GMM mean modeling and recent subspace techniques. Learn about applications in language and dialect recognition using deep neural networks.