Publication reference: https://doi.org/10.1007/978-981-16-1740-9_46This paper investigates the utilization of a physics simulation environment as the imagination of a robot, where it creates a replica of the detected terrain in a physics simulation environment in its memory, and “imagines” a simulated version of itself in that memory, performing actions and navigation on the terrain. The physics of the environment simulates the movement of robot parts and its interaction with the objects in the environment and the terrain, thus avoiding the need for explicitly programming many calculations.

Publication reference: https://doi.org/10.1007/978-981-99-9037-5_10 Sudoku puzzles are easily solved using backtracking algorithms. Yet, literature is scattered with various shy and often opaque attempts at using evolutionary algorithms and hybridizing them with known strategies of solving the puzzle. Evolutionary methods are serendipitous in nature, and this paper demonstrates the behaviour of such serendipity under constraints, using visuals that depict the sheer magnitude of the problem space and the nature in which intertwined constraints affect the scope for locating a solution, with the hope that it could inspire a new way of looking at the problem. We propose a method of visualizing the sudoku fitness landscape, the vastness and complexity of even partially brute forcing the puzzle, and a unique method of mutating puzzle states using circular swaps. These insights could potentially serve as a link to comprehend the problem space when designing solutions for vast, multidimensional problems. Additionally, finding the optimal solution for some puzzles was notably harder, compared to puzzles in the same category of given clues. A short investigation was conducted into this phenomenon, which revealed hints that compel us to propose that the direction of research that should be taken, is in discovering more about puzzle states and definitive mathematical properties of the puzzle, rather than merely designing brute-force, stochastic or hybrid approaches of finding solutions.

This research presents an empirical analysis on the realities of feature extraction and generalization of images of human faces. Experiments performed using principal component analysis (PCA) and histogram of oriented gradients (HOG) provided new insights on the variance of classification accuracy based on the quantity of data, and the actuality of data captured by specific principal components. HOG features provided greater classification accuracy for exact matches, and eigen faces of PCA were more accurate for generalization to previously unseen data. The results presented offer practical insights that can assist in the selection and design of algorithms that utilize PCA or HOG for data analysis. This is of particular significance in avoiding errors when data may undergo variations in sample size.

This research presents an empirical analysis on the realities of feature extraction and generalization of images of human faces. Experiments performed using principal component analysis (PCA) and histogram of oriented gradients (HOG) provided new insights on the variance of classification accuracy based on the quantity of data, and the actuality of data captured by specific principal components. HOG features provided greater classification accuracy for exact matches, and eigen faces of PCA were more accurate for generalization to previously unseen data. The results presented offer practical insights that can assist in the selection and design of algorithms that utilize PCA or HOG for data analysis. This is of particular significance in avoiding errors when data may undergo variations in sample size.

Any algorithm that needs to “understand” information to be capable of taking “intelligent” decisions, needs to access a lifetime of memories and experience the world as an embodied consciousness. This paper emphasizes these concepts and proposes a few fundamental constructs that provide algorithms with the capability to understand the human world, build larger sets of cooperative machines and perform causal inferences without requiring human intervention.

Emotion recognition by the human brain, normally incorporates context, body language, facial expressions, verbal cues, non-verbal cues, gestures and tone of voice. When considering only the face, piecing together various aspects of each facial feature is critical in identifying the emotion. Since viewing a single facial feature in isolation may result in inaccuracies, this paper attempts training neural networks to first identify specific facial features in isolation, and then use the general pattern of expressions on the face to identify the overall emotion. The reason for classification inaccuracies are also examined.

The design of any artificial intelligence is an interdisciplinary pursuit that requires keeping within perspective, the various properties of matter and life in the known universe, while remaining cautious of biases and misconceptions that arise from the limitations of prior learning, available tools and sensory capabilities. This paper curates a vast collection of human knowledge gathered during the process of exploring and questioning the fundamentals of why the mechanisms that constitute life were built in specific ways, and re-questioning those facts using anomalies that sometimes contradict or expose errors in human assumptions of life and intelligence. A meta-analysis of such knowledge at a single glance also helps identify interesting patterns in various phenomena that can in-time, spur creative solutions, alter the direction of research, assist with intelligent inferences and hopefully result in the identification and creation of artificial intelligence.