Capabilities currently classified as weak AI include successfully understanding human speech, competing at a high level in strategic game systems (such as chess and Go), self-driving cars, intelligent routing in content delivery networks, and interpreting complex data. Done through mainly calculus involving prediction problems, extrapolation based on feeding quality data.
The central problems (or goals) of AI research include reasoning, knowledge, planning, learning, natural language processing (communication), perception and the ability to move and manipulate objects.
General intelligence is among the field’s long-term goals. Approaches include statistical methods, computational intelligence, and traditional symbolic AI. Many tools are used in AI, including versions of search and mathematical optimization, logic, methods based on probability and economics. The AI field draws upon computer science, mathematics, psychology, linguistics, philosophy, neuroscience and artificial psychology.
From a deep learning perspective, even neuroevolution approaches has its limits for a strong AI, no matter how much the population of networks grow to attain the objective the acknowledgment of the solution has to have an external “decidant.”
“That is, evolution in this case is not just deciding the architecture and weights, but also the rules that guide how and when particular weights change.” from Neuroevolution: A different kind of deep learning by O.Stanley
In the context of strong AI, I really think that we need a kind of “attractors” framework as a “loosely” decidant role (aka. “conscience”).
“Strong AI is related to the field of consciousness, sentience, mind. We are on the way to general intelligence with every AI field evolving (from natural language processing to cognitive computing, social intelligence, planning, learning, perception etc.) The rules that govern the universe are already there, our mind works under the patterns within the universes pattern. To realize the strong AI we need to hook the machine into this flow of our as-such universe of small-world networks*. Consciousness is not programmable, it is a result of the sensory, when hooked to the flow, it can be seen as a result of the “strange attractors” created into the flow. A strong AI will exist when all the sensory inputs will have a proper cognitive processing, conscience will arise de facto from the space where the inputs exists but under a proactive threshold. The triggering occurs under strange attractors influenza, thus the conscience manifests itself.” (Definition of Strong AI by Essential-Works, C.Stefan 04.2017).
Neural network projects typically use some of the few architectural approaches like RNN, CNN, NTM, LSTM.
Deep learning architectures such as deep neural networks, deep belief networks and recurrent neural networks have been applied to fields including computer vision, speech recognition, natural language processing, audio recognition, social network filtering, machine translation and bioinformatics where they produced results comparable to and in some cases superior to human experts.
Neuroevolution is a subfield within artificial intelligence (AI) and machine learning (ML) that consists of trying to trigger an evolutionary process similar to the one that produced our brains, except inside a computer. In other words, neuroevolution seeks to develop the means of evolving neural networks through evolutionary algorithms.
Generative adversarial networks (GANs) are a class of neural networks more recently used in unsupervised machine learning. They facilitate a wide class of handy applications such as retrieving images that contain a given pattern or make predictions for a better medication to a certain disease.
APIs: Keras is a high-level neural networks API developed with a focus on enabling fast experimentation. Uses TensorFlow back-end engine by default.