3 Repos
Multi-task learning architectures that share bottom layers across multiple prediction objectives to learn from related tasks simultaneously.
Distinct from Multi-Task Learning Models: Distinct from Multi-Task Learning Models: specifically focuses on the shared-bottom architecture pattern rather than general multi-task sequence models.
Explore 3 awesome GitHub repositories matching artificial intelligence & ml · Shared-Bottom Architectures. Refine with filters or upvote what's useful.
DeepCTR is a specialized software framework and deep learning model library designed for predicting click-through rates and implementing recommendation systems. It provides a suite of tabular data models and architectures tailored for binary classification and sparse feature processing. The framework includes dedicated toolkits for multi-task learning and sequential interest modeling. It allows for the simultaneous estimation of multiple related targets through shared-bottom and gated expert neural networks, while capturing evolving user behavior using attention mechanisms and transformers.
Implements shared-bottom architectures to estimate multiple related targets using shared neural representations.
Sapiens ist ein hochauflösendes menschliches Sichtmodell, das für hochpräzise, menschenzentrierte Computer-Vision-Aufgaben entwickelt wurde. Es fungiert als Tool-Suite zur Schätzung menschlicher Posen, Tiefe und Oberflächengeometrie. Das Projekt nutzt ein Vision-Transformer-Backbone, um mehrere Aufgaben über einen gemeinsamen Encoder auszuführen. Diese Architektur ermöglicht die gleichzeitige Vorhersage von Skelettstrukturen, Gelenkpositionen und der Entfernung zwischen einer Kamera und einer menschlichen Person. Die Funktionen des Modells decken die Segmentierung menschlicher Körperteile zur Isolierung anatomischer Regionen vom Hintergrund sowie die Vorhersage von Oberflächennormalen zur Wiederherstellung von 3D-Geometriedetails aus 2D-Bildern ab. Diese Aufgaben werden durch ein Multi-Task-Learning-Framework unterstützt, das pixelweise Regression und semantische Segmentierungsmaskierung verwendet.
Uses a shared-bottom architecture to extract features for multiple specialized human-centric output heads.
DeepCTR-Torch is a deep learning library for building click-through rate prediction models. It provides a modular framework for assembling custom prediction architectures from pre-built core, interaction, and sequence layers, enabling the construction of deep neural networks that estimate click probability from user behavior data. The library specializes in feature interaction modeling, offering components for learning low-order, high-order, and adaptive-order feature crosses. It supports multi-task learning for predicting multiple objectives simultaneously, such as click and conversion rates
Shares bottom layers across multiple prediction objectives to learn click and conversion rates simultaneously.