Symbolic artificial intelligence Wikipedia

During the first AI summer, many people thought that machine intelligence could be achieved in just a few years. By the mid-1960s neither useful natural language translation systems nor autonomous tanks had been created, and a dramatic backlash set in. AI has the potential to revolutionize various industries by enabling machines to solve complex problems and think intuitively, going beyond mere automation. This encompasses various subfields and technologies, such as machine learning and natural language processing. Though statistical, deep learning models are now embedded in daily life, much of their decision process remains hidden from view. This lack of transparency makes it difficult to anticipate where the system is susceptible to manipulation, error, or bias.

Multiple different approaches to represent knowledge and then reason with those representations have been investigated. Below is a quick overview of approaches to knowledge representation and automated reasoning. International Standards for artificial intelligence provide a framework to guide the responsible and ethical use of AI technologies. These standards cover areas such as privacy, bias, transparency and accountability. By adhering to these standards, organizations can work to ensure that their AI systems are fair, transparent, and uphold ethical principles. For example, in basic machine learning, a computer could learn to recognize birds in pictures.

arXivLabs: experimental projects with community collaborators

Neither deep neural networks nor symbolic artificial intelligence (AI) alone has approached the kind of intelligence expressed in humans. This is mainly because neural networks are not able to decompose joint representations to obtain distinct objects (the so-called binding problem), while symbolic AI suffers from exhaustive rule searches, among other problems. These two problems are still pronounced in neuro-symbolic AI, which aims to combine the best of the two paradigms.

One example of International Standard in the AI field is ISO/IEC 23894, which focuses on the management of risk in AI systems. This standard aims to ensure that AI algorithms and models are understandable and can be audited for bias and fairness, thereby building trust and confidence in AI systems. With AI systems collecting vast amounts of data from databases worldwide, there is a need to ensure that personal information is protected and used responsibly. For example, facial recognition technology, often used in security systems or social media platforms, raises questions about consent and potential misuse. With an ability to synthesize, analyse and act on enormous amounts of data in seconds, artificial intelligence is extremely powerful. As with any powerful technology, it is crucial we implement it responsibly to maximize on its potential while minimizing negative impacts.

Neuro-symbolic AI research at the MIT-IBM Watson AI Lab

This lead towards the connectionist paradigm of AI, also called non-symbolic AI which gave rise to learning and neural network-based approaches to solve AI. How to explain the input-output behavior, or even inner activation states, of deep learning networks is a highly important line symbolic artificial intelligence of investigation, as the black-box character of existing systems hides system biases and generally fails to provide a rationale for decisions. Recently, awareness is growing that explanations should not only rely on raw system inputs but should reflect background knowledge.

However, both paradigms have strengths and weaknesses, and a significant challenge for the field today is to effect a reconciliation. A central tenet of the symbolic paradigm is that intelligence results from the manipulation of abstract compositional representations whose elements stand for objects and relations. If this is correct, then a key objective for deep learning is to develop architectures capable of discovering objects and relations in raw data, and learning how to represent them in ways that are useful for downstream processing. This short review highlights recent progress in this direction. Summarizing, neuro-symbolic artificial intelligence is an emerging subfield of AI that promises to favorably combine knowledge representation and deep learning in order to improve deep learning and to explain outputs of deep-learning-based systems.