This material may not be published, broadcast, rewritten, or redistributed. Quotes displayed in real-time or delayed by at least 15 minutes. Market data provided by Factset . Powered and implemented by FactSet Digital Solutions . Mutual Fund and ETF data provided by LSEG .

--We investigate the performance of a lightweight tracking controller, based on a flow version of the Newton-Raphson method, applied to a miniature blimp and a mid-size quadrotor . This tracking technique has been shown to enjoy theoretical guarantees of performance and has been applied with success in simulation studies and on mobile robots with simple motion models. This paper investigates the technique through real-world flight experiments on aerial hardware platforms subject to realistic deployment and onboard computational constraints. The technique's performance is assessed in comparison with the established control frameworks of feedback linearization for the blimp, and nonlinear model predictive control for both quadrotor and blimp. The performance metrics under consideration are (i) root mean square error of flight trajectories with respect to target trajectories, (ii) algorithms' computation times, and (iii) CPU energy consumption associated with the control algorithms. The experimental findings show that the Newton-Raphson flow-based tracking controller achieves comparable or superior tracking performance to the baseline methods with substantially reduced computation time and energy expenditure. HE past two decades have seen a significant shift in the nature of hardware research for trajectory control of aerial platforms like quadrotors. First, testing and verification of novel techniques relied heavily on numerical simulators, later transitioning to real-world deployments that depended on ground station computers and simplified models (e.g. Today, powerful single-board computers (SBCs) have enabled research to shift toward onboard execution even for computationally intensive control methods [2]-[4].

The massive increase in data in scientific research requires the development and application of robust tools for data analysis and m achine l earning (ML) that are findable, accessible, interoperable, re usable (FAIR) and interpretable. In domains, such as b iomaterials s cience, e ngineering, c hemistry, h ealthcare and b io sciences, data - driven discovery typically requires interdisciplinary teams . These teams collaborate to implement unbiased data pre - processing strategies, select appropriate modelling techniques, and interpret model outputs to accelerate and inform research outcomes and support rational design and decision - making. This process is often iterative, with experts providing feedback over long periods of time to refine models and optimise the methodology adopted . In cases where initial analysis identifies issues with the data, such as outliers, unbalance d data classes, or experimental measurement uncertainty, another round of data collection and pre - processing might be necessary . That means that data for the same problem are likely to be analysed multiple times using different dataset versions and methodological pipelines. For interdisciplinary co - development of analytic s, there is also a need for tools that allow domain experts to focus on interpreting and using analysis results, rather than developing code . The widespread use of ML and the overwhelming availability of thousands of community - driven open - source packages in Python and R increases the barrier for interoperable and reusable data analysis methodologies . To facilitate accurate analy tics, transparency, and modelling results comparison, there is a strong need for easy - to - use tools that automatically track data, all methodological choices, performance metrics, and corresponding results.

One of the fundamental problems within the field of machine learning is dimensionality reduction. Dimensionality reduction methods make it possible to combat the so-called curse of dimensionality, visualize high-dimensional data and, in general, improve the efficiency of storing and processing large data sets. One of the best-known nonlinear dimensionality reduction methods is Diffusion Maps. However, despite their virtues, both Diffusion Maps and many other manifold learning methods based on the spectral decomposition of kernel matrices have drawbacks such as the inability to apply them to data outside the initial set, their computational complexity, and high memory costs for large data sets. In this work, we propose to alleviate these problems by resorting to deep learning. Specifically, a new formulation of Diffusion Maps embedding is offered as a solution to a certain unconstrained minimization problem and, based on it, a cost function to train a neural network which computes Diffusion Maps embedding -- both inside and outside the training sample -- without the need to perform any spectral decomposition. The capabilities of this approach are compared on different data sets, both real and synthetic, with those of Diffusion Maps and the Nystrom method.

Mathematical reasoning is an increasingly important indicator of large language model (LLM) capabilities, yet we lack understanding of how LLMs process even simple mathematical tasks. To address this, we reverse engineer how three mid-sized LLMs compute addition. We first discover that numbers are represented in these LLMs as a generalized helix, which is strongly causally implicated for the tasks of addition and subtraction, and is also causally relevant for integer division, multiplication, and modular arithmetic. We then propose that LLMs compute addition by manipulating this generalized helix using the "Clock" algorithm: to solve $a+b$, the helices for $a$ and $b$ are manipulated to produce the $a+b$ answer helix which is then read out to model logits. We model influential MLP outputs, attention head outputs, and even individual neuron preactivations with these helices and verify our understanding with causal interventions. By demonstrating that LLMs represent numbers on a helix and manipulate this helix to perform addition, we present the first representation-level explanation of an LLM's mathematical capability.

Zane looks out to the calm ocean outside his window. This is his float time, when he lets his thoughts run wild. They generate a kaleidoscope, and he simply watches them, waiting to latch onto one that can be analyzed, deconstructed, and transformed into the next communication frequency. Zane exists in a world where humanity and artificial intelligence exist in a perfectly balanced symbiotic relationship. Human DNA is now intertwined with AI--a Triple Helix (TH)--embedded so deeply that ancestors from just a hundred years ago seem like mythical creatures.

This paper introduces Helix, a distributed system for high-throughput, low-latency large language model (LLM) serving on heterogeneous GPU clusters. A key idea behind Helix is to formulate inference computation of LLMs over heterogeneous GPUs and network connections as a max-flow problem for a directed, weighted graph, whose nodes represent GPU instances and edges capture both GPU and network heterogeneity through their capacities. Helix then uses a mixed integer linear programming (MILP) algorithm to discover highly optimized strategies to serve LLMs. This approach allows Helix to jointly optimize model placement and request scheduling, two highly entangled tasks in heterogeneous LLM serving. Our evaluation on several heterogeneous cluster settings ranging from 24 to 42 GPU nodes shows that Helix improves serving throughput by up to 2.7$\times$ and reduces prompting and decoding latency by up to 2.8$\times$ and 1.3$\times$, respectively, compared to best existing approaches.

Mechanistic Interpretability (MI) promises a path toward fully understanding how neural networks make their predictions. Prior work demonstrates that even when trained to perform simple arithmetic, models can implement a variety of algorithms (sometimes concurrently) depending on initialization and hyperparameters. Does this mean neuron-level interpretability techniques have limited applicability? We argue that high-dimensional neural networks can learn low-dimensional representations of their training data that are useful beyond simply making good predictions. Such representations can be understood through the mechanistic interpretability lens and provide insights that are surprisingly faithful to human-derived domain knowledge. This indicates that such approaches to interpretability can be useful for deriving a new understanding of a problem from models trained to solve it. As a case study, we extract nuclear physics concepts by studying models trained to reproduce nuclear data.

In recent years, the medical industry has witnessed a growing interest in minimally invasive procedures, with magnetic microrobots emerging as a promising approach. These micro-robots possess the ability to navigate through various media, including viscoelastic and non-Newtonian fluids, enabling targeted drug delivery and medical interventions. Many current designs, inspired by micro-swimmers in biological systems like bacteria and sperm, employ a contact-based method for transporting a payload. Adhesion between the cargo and the carrier can make release at the target site problematic. In this project, our primary objective was to explore the potential of a helical micro-robot for non-contact drug or cargo delivery. We conducted a comprehensive study on the shape and geometrical parameters of the helical microrobot, specifically focusing on its capability to transport passive filaments. Based on our analysis, we propose a novel design consisting of three sections with alternating handedness, including two pulling and one pushing microhelices, to enhance the capture and transport of passive filaments in Newtonian fluids using a non-contact approach. We then simulated the process of capturing and transporting the passive filament, and tested the functionality of the newly designed micro-robot. Our findings offer valuable insights into the physics of helical micro-robots and their potential for medical procedures and drug delivery. Furthermore, the proposed non-contact method for delivering filamentous cargo could lead to the development of more efficient and effective microrobots for medical applications.

Helix is a place where innovators and doers gather in order to drive significant progress in population genomics. We have come together to work at the intersection of clinical care, research, and genomics. If you're excited by the idea of making a meaningful impact and joining a team where we pride ourselves on driving innovation through fostering an environment with an emphasis on empowering one another to grow, Helix might be the place for you! Our end-to-end population genomics platform enables health systems, life sciences companies, and payers to advance genomic research and accelerate the integration of genomic data into routine clinical care. We support all aspects of population genomics from recruitment to translational research and help our partners use genomics to improve health outcomes, increase patient engagement, and lower costs.