Noah’s Latest Research
A Primer on the Challenges of Audio Latency in Artificial Intelligence Systems
Sep 5, 2024 | Artificial Intelligence
Audio latency in Artificial Intelligence (AI) systems poses significant challenges, especially in applications requiring real-time processing and interaction, such as the use of AI in call centers, translation, and live audio processing. This paper explores the technical complexities and mathematical frameworks underlying audio latency, including a brief analysis of its causes, impacts, and potential mitigation strategies. It aims to provide a comprehensive understanding of the challenges faced by AI systems in managing audio latency by looking at signal processing, neural network inference, and hardware-software co-design.
Policy Proposal for Improving SDG 4, SDG 8, and SDG 9
Nov 11, 2023 | Technology Policy
This policy proposal contains seven policies for secondary schools to increase college matriculation rates (quality education), increase workforce salaries, and improve industry, innovation, and infrastructure. Each policy aligns to one or more Sustainable Development Goals (SDGs) outlined by the United Nations.
Life vs. Intelligence: A Brief Comparative Study of Biological and Silicon Stacks in the AI Era
Sep 10, 2023 | Artificial Intelligence
Human society has long considered the biological stack to be the only life stack. However, the advent of silicon chips has forced us, or at least some of us, to contemplate the possibility of a second life stack – one composed of silicon instead of cells. Interestingly, this silicon stack is the life stack which humanity has come to fear, though the limitations of either life stack is still unknown. In this paper, we compare the biological and silicon life stacks (and whether “life stack” is the most appropriate terminology), with specific emphasis on the impact of Artificial Intelligence (AI) advancements on the silicon stack.
An Improved Neural Network Model Architecture
Jun 19, 2023 | Computational & Electrical Engineering
This research presents a new model for a neural network structure that is loosely based on the design of atoms and molecules, allowing for a circular flow of data. This has the benefit of allowing for pass-through nodes (increasing efficiency significantly) and allowing us to more accurately predict the binary state of a particular node in advance (0 or 1).
Analysis of Machine Learning and Computer Vision Emphasizing Unified Ant Colony Optimization
Mar 30, 2023 | Artificial Intelligence, Mathematics
Through this paper, we analyze specific branches of Artificial Intelligence (AI) technologies, such as Machine Learning (ML) and Computer Vision (CV). Specifically, we explore synergies between ML and CV, highlighting how Unified Ant Colony Optimization (ACO) serves as a unifying optimization technique in addressing complex problems within these domains. We also delve into the intricacies of ML and CV, examining both their theoretical foundations and practical applications, before exploring the Unified ACO algorithm and considering how it can be utilized to address difficult computational problems.
A Brief Analysis of the Architecture, Limitations, and Impacts of ChatGPT
Mar 23, 2023 | Artificial Intelligence
First, we begin with a description of the technical architecture of ChatGPT and
how it differs from other large scale artificial intelligence language models. From there, we define the current use cases of ChatGPT, followed by an analysis of some of the current limitations of ChatGPT. In particular, we look at the inability of ChatGPT to be creative, issues of perpetuating biases, and possibility of identification. Finally, we look at some of the likely key impacts of ChatGPT, including copyright considerations and economic ramifications. As much as possible, this analysis is done from a non-technical standpoint, and seeks to show how AI is beginning to connect with daily life.
Differentially Private Data and Data De- Identification
Mar 16, 2023 | Data Privacy & Security
This paper analyzes both differential privacy and data de-identification. While differential privacy seeks to create differentially private data through the use of mathematics, data de-identification seeks to anonymize data in such a way that it cannot be re-identified at a later date. In addition, we analyze the challenges of both methods of approaching privacy, including the possibility of data re-identification and verification of privacy, before addressing possible methods of mitigating these challenges. Such methods include setting outer bounds of data, utilizing shared central databases with larger datasets, and grouping data into fewer data category buckets. The merits and benefits of both methods are discussed as well.
An Analysis of Triangulation in Geometrically Noisy Environments using Mathematics
Feb 23, 2023 | Computational & Electrical Engineering, Mathematics
This paper uses mathematics to analyze the challenges of geometrically noisy environments on triangulation. Given widely accepted algorithmic triangulation methods, such as O (n ln n) or a simpler O (n3) method, we can mathematically prove that triangulation of any two dimensional polygonal region is possible, albeit impractical in some cases. Further, we consider the implications of environments in which a z-axis is present, as seen in cellular triangulation. In many of the cases where consideration of the z-axis is necessary, we recognize the absence of a fixed or known point of origin and consider methods of addressing this challenge.
An Analysis of the Increasing Processing Power of Modern Microprocessors
Dec 20, 2022 | Computational & Electrical Engineering
Traditional computational power calculations rely on the assumption that additional processing power is best achieved by a quantitative increase in transistors, and particularly through a quantitative increase in the number of metal-oxide-semiconductor field-effect transistors (MOSFETs). However, advancements in microprocessor architecture could fundamentally alter the method in which additional processing power is achieved, thus rendering traditional computational power calculations meritless.