Neural Networks - Movie Review Example

Neural Networks The movie, exmachina explores the use of robots to perform tasks that seem to be beyond the human’s capabilities. This is achieved through ‘teaching’ the robot in a form of machine learning and implementing the cognitive abilities of humans into the robot but with the high levels inventory of course. The robots are therefore made adaptive as much as possible so that they can accept varied inputs and learn faster with ease since they develop higher recognition levels for patterns, sounds or even light intensities.

With neural networks the robot are able to check their payload; the load capacity they can contain in any operation and there be able to know the maximum load beyond which it fails or develop dramatic loss in accuracy. In the story, it is very clear that through neural networks robots can be controlled remotely through gestures and be directed to perform some tasks depending their interpretation of those gestures or symbols shown. Neural networks have been used for robot control over long time now and have evolved to high levels of control that are intricate.

Firstly, a robot can be learn through the neural networks; this is achieved a neural network brain for the robot. An example of a good robot learning task is navigational tasks that may include learning to detect objects or obstacles like a wall and to move about a space in a defined path. Through the learning, the robots can develop maps in their ‘brains’ for the environment they are in after being taught. This includes recognizing as well as being able to predict behaviors and upcoming obstacles on their paths through voice recognition and may be light detecting sensors within them.

Critical decision-making can be inculcated in robots by neural networks where it is supposed to weigh the negative and positive effects of its actions depending the outcomes of the action it is about to take. This shows the analytical characteristics involved and measurement of the levels of effects caused by the activities undertaken by the robots. For example, given more than one task to perform which have different slack times to be finished on their due dates, the robot will determine the best option by choosing the task with the least slack times to be performed first.

Another area where the neural networks have been used is in the control of the robots movements in the joints as it performs it tasks. It can offer a continuous movement or a discontinuous movement that is stepped. Depending on the pace of the tasks it is expected to perform the robot is programmed and with the ability of the neural network within it to control its actuators that are interlinked, it is able to either move continuously with a slow motion on the joints or a provide a jerking motion of stepping.

Some among the improvements that I would recommend include; multimodal integration of robot behaviors using deep neural networks which is more advanced. This is because it enhances perceptual precision and reduces ambiguity in the robot recognition ability. The sensory information from most of the sources such as vision, audition, and proprioception are combined, hence an enhanced perceptual clarity and reduced ambiguity regarding the robots environment. Through the incorporation of Boolean concepts within the neural system of the robot would enable it making more rational decisions and aid in good analytical problem-solving.

The Boolean statements are programmed in the robot memory alongside the tasks to be performed, therefore depending on the tasks chosen; all this would be after a thorough check on the right sequence to be followed by the robot. For a sensor-based controlled robot with vision, I would use real-time application of neural networks to improve the feedback mechanism from the sensors and, therefore, generate pulses for actions faster. One network in the controller is used to learn to reproduce the nonlinear relationship between the sensor outputs and the robots system command variables over a particular regions of the systems state space.

The learned information is used to predict the command signals required to produce desired changes in the sensor outputs. The second network is used to learn to reproduce the nonlinear relationship between the system command variables and the changes in the video sensor outputs. The learned information from this network is then used to predict the next set of video parameters, effectively compensating for the image processing delays. The movie was enjoyable, and a fantastic experience since the various applications of science fiction were evident with a clear indication of the end results of the robots actions in accordance with the demands it is fed with.

In addition in levels interaction between the robot and humans it is also portrayed in a manner to bring out a comparison between the artificial intelligence among the robots and the humans that develop them.Referencehttp://exmachina-movie.com/