C. Daniel Freeman
YOU?
Author Swipe
View article: Mobile language app learners’ self-efficacy increases after using generative AI
Mobile language app learners’ self-efficacy increases after using generative AI Open
Introduction Although generative artificial intelligence (AI) is ubiquitous, there is little research on how it supports self-efficacy (learners’ belief that they can perform at a particular level on a specific task). The purpose of these …
View article: Learned Neural Physics Simulation for Articulated 3D Human Pose Reconstruction
Learned Neural Physics Simulation for Articulated 3D Human Pose Reconstruction Open
We propose a novel neural network approach, LARP (Learned Articulated Rigid body Physics), to model the dynamics of articulated human motion with contact. Our goal is to develop a faster and more convenient methodological alternative to tr…
View article: Training Language Models on the Knowledge Graph: Insights on Hallucinations and Their Detectability
Training Language Models on the Knowledge Graph: Insights on Hallucinations and Their Detectability Open
While many capabilities of language models (LMs) improve with increased training budget, the influence of scale on hallucinations is not yet fully understood. Hallucinations come in many forms, and there is no universally accepted definiti…
View article: Commensal-derived short-chain fatty acids disrupt lipid membrane homeostasis in<i>Staphylococcus aureus</i>
Commensal-derived short-chain fatty acids disrupt lipid membrane homeostasis in<i>Staphylococcus aureus</i> Open
The role of commensal anaerobic bacteria in chronic respiratory infections is unclear, yet they can exist in abundances comparable to canonical pathogens in vivo . Their contributions to the metabolic landscape of the host environment may …
View article: Defective <i>pgsA</i> contributes to increased membrane fluidity and cell wall thickening in <i>Staphylococcus aureus</i> with high-level daptomycin resistance
Defective <i>pgsA</i> contributes to increased membrane fluidity and cell wall thickening in <i>Staphylococcus aureus</i> with high-level daptomycin resistance Open
Daptomycin is a membrane-targeting last-resort antimicrobial therapeutic for the treatment of infections caused by methicillin- and/or vancomycin-resistant Staphylococcus aureus . In the rare event of failed daptomycin therapy, the source …
View article: Frontier Language Models are not Robust to Adversarial Arithmetic, or "What do I need to say so you agree 2+2=5?
Frontier Language Models are not Robust to Adversarial Arithmetic, or "What do I need to say so you agree 2+2=5? Open
We introduce and study the problem of adversarial arithmetic, which provides a simple yet challenging testbed for language model alignment. This problem is comprised of arithmetic questions posed in natural language, with an arbitrary adve…
View article: Improving Large Language Model Fine-tuning for Solving Math Problems
Improving Large Language Model Fine-tuning for Solving Math Problems Open
Despite their success in many natural language tasks, solving math problems remains a significant challenge for large language models (LLMs). A large gap exists between LLMs' pass-at-one and pass-at-N performance in solving math problems, …
View article: Defective pgsA contributes to increased membrane fluidity and cell wall thickening in S. aureus with high-level daptomycin resistance
Defective pgsA contributes to increased membrane fluidity and cell wall thickening in S. aureus with high-level daptomycin resistance Open
Daptomycin is a membrane-targeting last-resort antimicrobial therapeutic for the treatment of infections caused by methicillin- and/or vancomycin-resistant Staphylococcus aureus . In the rare event of failed daptomycin therapy, the source …
View article: Transformer-Based Learned Optimization
Transformer-Based Learned Optimization Open
We propose a new approach to learned optimization where we represent the computation of an optimizer's update step using a neural network. The parameters of the optimizer are then learned by training on a set of optimization tasks with the…
View article: VeLO: Training Versatile Learned Optimizers by Scaling Up
VeLO: Training Versatile Learned Optimizers by Scaling Up Open
While deep learning models have replaced hand-designed features across many domains, these models are still trained with hand-designed optimizers. In this work, we leverage the same scaling approach behind the success of deep learning to l…
View article: Allowing Safe Contact in Robotic Goal-Reaching: Planning and Tracking in Operational and Null Spaces
Allowing Safe Contact in Robotic Goal-Reaching: Planning and Tracking in Operational and Null Spaces Open
In recent years, impressive results have been achieved in robotic manipulation. While many efforts focus on generating collision-free reference signals, few allow safe contact between the robot bodies and the environment. However, in human…
View article: Practical tradeoffs between memory, compute, and performance in learned optimizers
Practical tradeoffs between memory, compute, and performance in learned optimizers Open
Optimization plays a costly and crucial role in developing machine learning systems. In learned optimizers, the few hyperparameters of commonly used hand-designed optimizers, e.g. Adam or SGD, are replaced with flexible parametric function…
View article: Gradients are Not All You Need
Gradients are Not All You Need Open
Differentiable programming techniques are widely used in the community and are responsible for the machine learning renaissance of the past several decades. While these methods are powerful, they have limits. In this short report, we discu…
View article: Brax -- A Differentiable Physics Engine for Large Scale Rigid Body Simulation
Brax -- A Differentiable Physics Engine for Large Scale Rigid Body Simulation Open
We present Brax, an open source library for rigid body simulation with a focus on performance and parallelism on accelerators, written in JAX. We present results on a suite of tasks inspired by the existing reinforcement learning literatur…
View article: Revealing Fatty Acid Heterogeneity in Staphylococcal Lipids with Isotope Labeling and RPLC–IM–MS
Revealing Fatty Acid Heterogeneity in Staphylococcal Lipids with Isotope Labeling and RPLC–IM–MS Open
Up to 80% of the fatty acids in Staphylococcus aureus membrane lipids are branched, rather than straight-chain, fatty acids. The branched fatty acids (BCFAs) may have either an even or odd number of carbons, and the branch position may be …
View article: Training Learned Optimizers with Randomly Initialized Learned Optimizers
Training Learned Optimizers with Randomly Initialized Learned Optimizers Open
Learned optimizers are increasingly effective, with performance exceeding that of hand designed optimizers such as Adam~\citep{kingma2014adam} on specific tasks \citep{metz2019understanding}. Despite the potential gains available, in curre…
View article: Tasks, stability, architecture, and compute: Training more effective\n learned optimizers, and using them to train themselves
Tasks, stability, architecture, and compute: Training more effective\n learned optimizers, and using them to train themselves Open
Much as replacing hand-designed features with learned functions has\nrevolutionized how we solve perceptual tasks, we believe learned algorithms\nwill transform how we train models. In this work we focus on general-purpose\nlearned optimiz…
View article: Tasks, stability, architecture, and compute: Training more effective learned optimizers, and using them to train themselves
Tasks, stability, architecture, and compute: Training more effective learned optimizers, and using them to train themselves Open
Much as replacing hand-designed features with learned functions has revolutionized how we solve perceptual tasks, we believe learned algorithms will transform how we train models. In this work we focus on general-purpose learned optimizers…
View article: Modern Approaches to Exact Diagonalization and Selected Configuration Interaction with the Adaptive Sampling CI Method
Modern Approaches to Exact Diagonalization and Selected Configuration Interaction with the Adaptive Sampling CI Method Open
Recent advances in selected configuration interaction methods have made them competitive with the most accurate techniques available and, hence, creating an increasingly powerful tool for solving quantum Hamiltonians. In this work, we buil…
View article: Using a thousand optimization tasks to learn hyperparameter search strategies
Using a thousand optimization tasks to learn hyperparameter search strategies Open
We present TaskSet, a dataset of tasks for use in training and evaluating optimizers. TaskSet is unique in its size and diversity, containing over a thousand tasks ranging from image classification with fully connected or convolutional neu…
View article: Learning to Predict Without Looking Ahead: World Models Without Forward Prediction
Learning to Predict Without Looking Ahead: World Models Without Forward Prediction Open
Much of model-based reinforcement learning involves learning a model of an agent's world, and training an agent to leverage this model to perform a task more efficiently. While these models are demonstrably useful for agents, every natural…
View article: Understanding and correcting pathologies in the training of learned optimizers
Understanding and correcting pathologies in the training of learned optimizers Open
Deep learning has shown that learned functions can dramatically outperform hand-designed functions on perceptual tasks. Analogously, this suggests that learned optimizers may similarly outperform current hand-designed optimizers, especiall…
View article: Understanding and correcting pathologies in the training of learned\n optimizers
Understanding and correcting pathologies in the training of learned\n optimizers Open
Deep learning has shown that learned functions can dramatically outperform\nhand-designed functions on perceptual tasks. Analogously, this suggests that\nlearned optimizers may similarly outperform current hand-designed optimizers,\nespeci…
View article: Stable quantum memories with limited measurement
Stable quantum memories with limited measurement Open
We demonstrate the existence of a finite temperature threshold for a 1D\nstabilizer code under an error correcting protocol that requires only a\nfraction of the syndrome measurements. Below the threshold temperature, encoded\nstates have …
View article: Nuclear Magnetic Resonance Spectroscopy Investigations of Naphthalene-Based 1,2,3-Triazole Systems for Anion Sensing
Nuclear Magnetic Resonance Spectroscopy Investigations of Naphthalene-Based 1,2,3-Triazole Systems for Anion Sensing Open
Detailed Nuclear Magnetic Resonance (NMR) spectroscopy investigations on a novel naphthalene-substituted 1,2,3-triazole-based fluorescence sensor provided evidence for the “turn-on” detection of anions. The one-step, facile synthesis of th…
View article: The Toric Code at Finite Temperature
The Toric Code at Finite Temperature Open
Alexei Kitaev's toric code is a rich model, that has birthed and stimulated the development of topological quantum computing, error correction, and field theory. It was also the first example of a quantum error correcting code that could r…
View article: CCDC 1582561: Experimental Crystal Structure Determination
CCDC 1582561: Experimental Crystal Structure Determination Open
An entry from the Cambridge Structural Database, the world’s repository for small molecule crystal structures. The entry contains experimental data from a crystal diffraction study. The deposited dataset for this entry is freely available …
View article: Monte Carlo Tensor Network Renormalization
Monte Carlo Tensor Network Renormalization Open
Techniques for approximately contracting tensor networks are limited in how efficiently they can make use of parallel computing resources. In this work we demonstrate and characterize a Monte Carlo approach to the tensor network renormaliz…