Boosted Trees and Deep Neural Networks for Better Recommender Systems
Explore GPU-accelerated boosted trees and deep neural networks for building winning recommender systems, featuring insights from Kaggle Grandmasters on the ACM RecSys Challenge.
Machine Learning at the Extreme Edge of Always-on Intelligent Sensor Networks
Explore ML optimization techniques for resource-constrained IoT sensor nodes, focusing on system, algorithm, architecture, and circuit levels to achieve significant area-power improvements.
FedLess - Secure and Scalable Serverless Federated Learning
Explore serverless federated learning with FedLess, addressing scalability and security challenges in collaborative model training across distributed clients using cloud platforms.
TinyRL: Teaching AI to Swing Up a Real Pendulum
Explore reinforcement learning to train an AI agent to swing up a real pendulum using Arduino, Python frameworks, and ESP32 deployment, with practical demonstrations and troubleshooting tips.
A Decade of Machine Learning Accelerators: Lessons Learned and Carbon Footprint
Exploring a decade of machine learning accelerators, lessons learned, and strategies to reduce the carbon footprint of deep neural networks through optimizing models, machines, mechanisms, and mapping.
Domain Specific Architectures for Deep Neural Networks: Three Generations of Tensor Processing Units
Exploring three generations of Tensor Processing Units (TPUs) for deep neural networks, highlighting performance improvements and the rise of domain-specific architectures in the post-Moore's Law era.
Sustainable Success - AI's Impact on Humanity and Innovation
Explore AI's transformative impact on humanity with Dr. Srinivas Kishan, a double PhD holder in AI. Discover how this technology can empower us to tackle complex problems and redefine possibilities.
The Softmax Function, Potts Model and Variational Neural Networks in Image Segmentation
Exploring variational models as layers in deep neural networks for image segmentation, integrating traditional techniques with DNNs to improve performance and incorporate shape priors.
Parcae: Proactive, Liveput-Optimized DNN Training on Preemptible Instances
Optimize DNN training costs using preemptible cloud instances with Parcae, a system that proactively adjusts parallelization strategies based on predicted resource changes, improving performance and scalability.
News on Tensor Networks for Machine Learning and Quantum Computing Simulation
Explore recent advancements in tensor networks for enhancing machine learning and simulating quantum processors, covering deep neural networks, convolutional networks, and large language models.
Data-Driven Regularisation for Solving Inverse Problems - Carola-Bibiane Schönlieb, Turing/Cambridge
Explores data-driven regularizers for inverse imaging problems, combining model-based and data-driven approaches. Covers bilevel optimization and deep neural networks for image processing.
Taming Reachability Analysis of DNN-Controlled Systems via Abstraction-Based Training
Explore a novel abstraction-based approach for efficient reachability analysis of DNN-controlled systems, improving tightness and efficiency over existing methods.
I-Vector Representation Based on GMM and DNN for Audio Classification
Explore advanced I-vector techniques for audio classification, including GMM and DNN-based approaches. Learn about subspace modeling and applications in language and dialect recognition.
Explainable AI with Applications for Brain Imaging and Drug Discovery
Explore explainable AI techniques for deep neural networks, focusing on saliency backpropagation. Learn applications in biomarker discovery from fMRI data and drug mechanism prediction.
Principles and Practice of Scalable and Distributed Deep Neural Networks Training and Inference
Dive into scalable and distributed deep neural networks, exploring advanced training and inference techniques for enhanced performance and efficiency in modern AI systems.
Static Detection of Novel Malware Using Transfer Learning with Deep Neural Networks
Explore innovative malware detection techniques using transfer learning and deep neural networks for enhanced cybersecurity and threat prevention.
AI - Hacking without Humans - How Can Human Brains Be Hacked?
Explore the security, ethical, and privacy concerns of AI models mimicking human brain functions. Learn about adversarial AI, intelligence sharing, and building strong defenses in cybersecurity.
Machine Learning & Artificial Intelligence
Explore machine learning and AI's impact on education and life outcomes with neuroscientist Vivienne Ming. Discover innovative applications of technology for maximizing human potential.
Artificial Intelligence
Explore AI's impact on jobs, creativity, and human potential with Dr. Vivienne Ming. Discover how augmented intelligence and machine learning can shape a better future and address global challenges.
Machine Duping - Pwning Deep Learning Systems
Explore vulnerabilities in deep learning systems and learn techniques to manipulate image recognition, speech recognition, and phishing detection through practical demonstrations and attack methodologies.
A Sound Mind in a Vulnerable Body - Practical Hardware Attacks on Deep Learning
Explores hardware vulnerabilities in deep learning systems, demonstrating how fault injection and side-channel attacks can compromise model accuracy and steal proprietary information, highlighting the need for robust ML-level defenses.