Computer Vision for Segmenting & Analyzing Subcellular Components in Cryo-Electron Tomography
Explore computer vision techniques for automated 3D segmentation and analysis of subcellular components in cryo-electron tomography, focusing on mitochondrial features in Huntington's Disease neurons.
Learning to Explore with Scalable Supervision - IPAM at UCLA
Discover efficient reinforcement learning techniques for robotic control, balancing self-supervision and human guidance to overcome exploration challenges in high-dimensional spaces.
Multiview and Self-Supervised Representation Learning - Nonlinear Mixture Identification
Explore nonlinear mixture identification in multiview and self-supervised representation learning. Gain insights into extracting essential information from high-dimensional data for improved AI applications.
Enabling On-Device Learning at Scale
Explore on-device learning techniques for AI personalization, including few-shot learning, continuous learning, and federated learning, addressing privacy and efficiency challenges in edge computing.
Audio-Visual Self-Supervised Learning: Insights from Infant Development
Explore how neural networks learn audio-visual representations from video streams, mimicking infant learning through synchronization of sight and sound.
Recent Advances in Foundation Models: Scaling Laws, Emergent Behaviors, and AI Democratization
Explore recent advances in foundation models, scaling laws, and emergent behaviors in AI. Learn about efforts to democratize AI and make cutting-edge technology accessible to all.
Emu Video Generation - From MAE Pre-training to Multimodal Embeddings
Explore Emu Video generation with Ishan Misra from Meta, covering SSL, multimodal data, and recent advancements in video AI technology.
Bridging ML Models with Medical Insights for Learning, Inference, and Model Explanation in Radiology
Explore ML models bridging medical insights for radiology applications. Learn about self-supervised representation, progressive exaggeration, contrastive learning, causal analysis, and association tests in medical imaging.
Building Better Language Models - Paradigms and Techniques
Explore advanced techniques for building and improving language models, covering paradigms, zero-shot learning, and multitask prompted training.
NYU Deep Learning
Comprehensive deep learning course covering neural networks, convolutional and recurrent architectures, energy-based models, self-supervised learning, and advanced topics like transformers and graph networks.
GeeSolver - A Generic, Efficient, and Effortless Solver with Self-Supervised Learning for Breaking Text Captchas
Explore GeeSolver, a cutting-edge self-supervised learning approach for efficiently breaking text CAPTCHAs, offering insights into advanced cybersecurity techniques and challenges.
BadEncoder - Backdoor Attacks to Pre-trained Encoders in Self-Supervised Learning
Explore backdoor attacks on pre-trained encoders in self-supervised learning, examining vulnerabilities and potential security risks in AI systems.
Self-Supervised Learning - Towards Autonomously Learning Machines
Explore cutting-edge AI techniques for autonomous learning machines, focusing on self-supervised learning methods that reduce reliance on labeled data and enhance general understanding.
The Capabilities of Semantic Search
Explore semantic search capabilities with the Jina framework, covering symbolic vs. semantic approaches and practical applications for various data types.
Pre-training Molecular Graph Representation with 3D Geometry
Explore molecular graph representation learning, leveraging 3D geometry for enhanced drug and material discovery through self-supervised learning and the GraphMVP framework.
UnSAM - Unsupervised Segmentation Anything Model: A New Approach to Image Segmentation
Discover UnSAM: a groundbreaking unsupervised image segmentation model. Learn its divide-and-conquer approach, hierarchical structure discovery, and impressive performance across datasets.
Towards Physically Reliable Molecular Representation Learning - UAI 2023 Oral Session 5
Explore innovative molecular representation learning for estimating energetic properties, combining physics-driven parameter estimation and self-supervised learning to enhance extrapolation beyond training data.
The Rise of Neural Priors
Explore neural priors as a solution to AI challenges, including data scarcity and bias. Learn about their applications in augmented reality, autonomous driving, and signal recovery.
Big Self-Supervised Models Are Strong Semi-Supervised Learners
Explore SimCLRv2, a powerful semi-supervised learning approach combining self-supervised pre-training, fine-tuning, and distillation to achieve state-of-the-art results with minimal labeled data.
Bootstrap Your Own Latent: A New Approach to Self-Supervised Learning - Paper Explained
Explore BYOL, a novel self-supervised learning approach that outperforms baselines without negative samples. Learn about its innovative use of online and target networks for image representation.
Towards Better Understanding of Contrastive Learning
Explore contrastive learning's foundations, dynamics, and applications in deep models. Gain insights into self-supervised learning, nonlinear analysis, and feature emergence through expert presentation.