We describe algorithms and data structures to extend a neural network library with automatic precision estimation for floating point computations. We also discuss conditions to make estimations exact and preserve high computation performance of neural networks training and inference. Numerical experiments show the consequences of significant precision loss for particular values such as inference, gradients and deviations from mathematically predicted behavior. It turns out that almost any neural network accumulates computational inaccuracies. As a result, its behavior does not coincide with predicted by the mathematical model of neural network. This shows that tracking of computational inaccuracies is important for reliability of inference, training and interpretability of results.

In this thesis, we aim to improve the performance of TAMP algorithms from three complementary perspectives. First, we investigate the integration of discrete task planning with continuous trajectory optimization. Our main contribution is a conflict-based solver that automatically discovers why a task plan might fail when considering the constraints of the physical world. This information is then fed back into the task planner, resulting in an efficient, bidirectional, and intuitive interface between task and motion, capable of solving TAMP problems with multiple objects, robots, and tight physical constraints. In the second part, we first illustrate that, given the wide range of tasks and environments within TAMP, neither sampling nor optimization is superior in all settings. To combine the strengths of both approaches, we have designed meta-solvers for TAMP, adaptive solvers that automatically select which algorithms and computations to use and how to best decompose each problem to find a solution faster. In the third part, we combine deep learning architectures with model-based reasoning to accelerate computations within our TAMP solver. Specifically, we target infeasibility detection and nonlinear optimization, focusing on generalization, accuracy, compute time, and data efficiency. At the core of our contributions is a refined, factored representation of the trajectory optimization problems inside TAMP. This structure not only facilitates more efficient planning, encoding of geometric infeasibility, and meta-reasoning but also provides better generalization in neural architectures.

FPGAs' ability to distribute massive workloads into parallel computation enables AI features to create highly efficient electronic devices. FREMONT, CA: Implementing FPGA increases the number of parallel computational elements and processing efficiency of the electronic devices. FPGAs that hold parallel and hardware-programmable feature enables electronics excellence at specialized workloads with high computational operations and optimal configurations. Over the past few years, FPGAs have proved to be the low-power solution, making it flexible and ideal for Neural Network (NN) architectures. Today, professionals are focusing on creating designs for supporting AI-based applications and functions.

Artificial Intelligence (AI) is probably one of the branches of Computer Science that is experiencing more growth nowadays. Even though it was born more than 70 years ago, it is in a historical period where it has generated more interest because of the revolution it has caused on the market today. Until very recently, there were limited computational capacities that made Artificial Intelligence produce very poor results on the problem being applied, which resulted in several periods of historical dissatisfaction in the industry and considerable reduction in both interests. In this case, discipline on the number of dedicated researchers. However, in recent years Artificial Intelligence has gained enormous momentum, able to solve problems with computers that were previously thought to be impossible, reaching levels that had never been reached before.

A Computer-Assisted Reading and Analysis of Texts (CARAT) process is a complex technology that connects language, text, information and knowledge theories with computational formalizations, statistical approaches, symbolic approaches, standard and non-standard logics, etc. This process should be, always, under the control of the user according to his subjectivity, his knowledge and the purpose of his analysis. It becomes important to design platforms to support the design of CARAT tools, their management, their adaptation to new needs and the experiments. Even, in the last years, several platforms for digging data, including textual data have emerged; they lack flexibility and sound formal foundations. We propose, in this paper, a formal model with strong logical foundations, based on typed applicative systems.

Most neuroscientists believe that the brain learns by rewiring itself--by changing the strength of connections between brain cells, or neurons. But experimental results published a few years ago, from a lab at Lund University in Sweden, hint that we need to change our approach. They suggest the brain learns in a way more analogous to that of a computer: It encodes information into molecules inside neurons and reads out that information for use in computational operations. With a computer scientist, Adam King, I co-authored a book, Memory and the Computational Brain: Why Cognitive Science Will Transform Neuroscience. We argued that well-established results in cognitive science and computer science imply that computation in the brain must resemble computation in a computer in just this way.

Consider observation data, comprised of n observation vectors with values on a set of attributes. This gives us n points in attribute space. Having data structured as a tree, implied by having our observations embedded in an ultrametric topology, offers great advantage for proximity searching. If we have preprocessed data through such an embedding, then an observation's nearest neighbor is found in constant computational time, i.e. O(1) time. A further powerful approach is discussed in this work: the inducing of a hierarchy, and hence a tree, in linear computational time, i.e. O(n) time for n observations. It is with such a basis for proximity search and best match that we can address the burgeoning problems of processing very large, and possibly also very high dimensional, data sets.