Category
2,295 papers
This paper proposes a novel framework for constructing and evolving a neuro-symbolic knowledge graph (NSKG) capable of adapting to dynamic changes in information. The core idea is to combine the strengths of deep learning and knowledge graphs, leveraging graph neural networks (GNNs) to learn relationships within the gr...
Reinforcement Learning (RL) has demonstrated remarkable success in various domains, but its application is significantly hindered by the problem of sparse rewards. In environments where rewards are infrequent or delayed, traditional RL algorithms struggle to learn effective policies. This paper proposes a novel approac...
Graph embeddings have emerged as a powerful technique for representing complex networks in a low-dimensional vector space. However, traditional graph embedding methods often lack interpretability, making it difficult to understand the underlying reasons for the generated embeddings. This paper proposes a novel dynamic...
This paper introduces a novel approach to enhancing Graph Neural Networks (GNNs) by incorporating topological information. The core idea is to explicitly model and learn the relationships between nodes and edges within a graph, leading to improved representation learning and training efficiency, particularly in complex...
Reach audiences
Advertise in front of researchers, engineers, and readers.
This paper proposes a novel approach to neuro-symbolic programming that leverages a dynamic knowledge graph construction process. The core idea is to integrate neural networks with symbolic programming, enhancing the system's ability to learn and reason. A neural network is trained to generate symbolic rules and assert...
This paper explores the application of Graph Neural Networks (GNNs) to program structure optimization. Traditional program optimization techniques often rely on static analysis and hand-crafted rules, which can be limited in their ability to capture complex code dependencies and adapt to diverse program styles. We prop...
This paper presents a novel approach to intelligent perception systems by leveraging multi-scale information fusion. The core idea is to effectively integrate diverse data modalities – including low-resolution, high-resolution images, audio, and text – represented through a graph-based framework. This framework utilize...
This paper introduces the Dynamic Semantic Embedding Network (DSEN), a novel approach to program understanding and prediction that leverages the power of graph neural networks and continuous learning. The core claim of DSEN is to achieve dynamic comprehension and prediction of program behavior without explicit annotati...
Knowledge graph completion (KGC) is a critical task in leveraging the rich structural information embedded within knowledge graphs (KGs). Traditional graph neural networks (GNNs) often face challenges in effectively capturing the intricate relationships and dependencies within these graphs, particularly when dealing wi...
This paper proposes a novel approach to semantic understanding utilizing multi-scale graph neural networks (MS-GNNs). Traditional semantic understanding models often operate within a single scale of representation, limiting their ability to capture nuanced relationships and contextual information. Our method addresses...
This paper introduces a novel approach to knowledge graph reasoning based on Dynamic Graph Neural Networks (DGNNs). Traditional knowledge graph reasoning models often treat knowledge graphs as static structures, failing to capture the inherent dynamic evolution of entities and relationships. This work addresses this li...
This paper presents a novel approach to verifying complex temporal logic specifications on dynamic systems represented as graphs, addressing the significant computational challenges inherent in traditional verification methods. The core idea is to leverage the scalability of Probabilistic Neural Networks (PNNs) to appr...
A cross-section of the MIT community convened discussions on topics ranging from data centers and super pollutants to agriculture and education.
Machine-learning networks can learn without detailed instructions from a designer. Those that do may offer insight into how the brain reorganizes itself during learning.
Assistant Professor Pat Pataranutaporn describes a new interface that lets everyday users glimpse inside an AI's neural network before their chatbot ever says a word.
Cryptographic code supports vital protections in modern computing systems. Learn how a new method helps verify code as developers write it while preserving speed and adaptability as it gets implemented and evolves. The post Verifying Rust cryptography in SymCrypt, from standards to code appeared first on Microsoft Research.
PhD student Rachel Sava, winner of the Envisioning the Future of Computing Prize, explores transformative improvements and dystopian risks of neural technology.
Climate & Sustainability
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
