arXiv subjects

Daniel Huang

Publications and source records attributed to Daniel Huang.

15 recordsLinked to original sources

arXiv2026

StatQAT: Statistical Quantizer Optimization for Deep Networks

Quantization is essential for reducing the computational cost and memory usage of deep neural networks, enabling efficient inference on low-precision hardware. Despite the growing adoption of uniform and floating-point quantization schemes, selecting optimal quantization parameters remains a key challenge, particularly for diverse data distributions encountered during training and inference. This work presents a novel statistical error analysis framework for uniform and floating-point quantization, providing theoretical insight into error behavior across quantization configurations. Building on this analysis, we propose iterative quantizers designed for arbitrary data distributions and analytic quantizers tailored for Gaussian-like weight distributions. These methods enable efficient, low-error quantization suitable for both activations and weights. We incorporate our quantizers into quantization-aware training and evaluate them across integer and floating-point formats. Experiments demonstrate improved accuracy and stability, highlighting the effectiveness of our approach for training low-precision neural networks.

stat.ML↗

arXiv2025

What Is Your Agent's GPA? A Framework for Evaluating Agent Goal-Plan-Action Alignment

We introduce the Agent GPA (Goal-Plan-Action) framework, driven by the fundamental insight that critical agent failures emerge at the intersections of setting goals, devising plans, and executing actions. We operationalize the framework with a factorized suite of LLM judges designed to measure distinct elements of Goal-Plan-Act alignment. To make this methodology scalable and generalizable across diverse agent architectures and datasets, we use state-of-the-art automated prompt optimization techniques to systematically generate domain-specific evaluation criteria. We validate this approach across three benchmarks: a multi-agent research setting (TRAIL/GAIA), a single coding agent setting (TRAIL/SWE-bench), and a private, enterprise data-agent setting (Snowflake Intelligence). Extensive evaluation on TRAIL/GAIA demonstrates the core validity of the framework, which identifies a broad range of agent failures (95% of human-annotated errors), localizes errors to enable targeted debugging (86% of human-annotated errors), and exhibits strong agreement with human evaluators. Crucially, by applying our automated methodology to both public datasets, we demonstrate that our GPA judges generally achieve the highest error coverage (ranging from 76% to 86%) in comparison to manual prompting approaches. We also leverage an evolutionary coding agent to improve judge consistency by up to 38% through iterative refinement of evaluation rubrics. Overall, Agent GPA provides a rigorous and generalizable paradigm for targeted agent evaluation.

cs.AI↗

arXiv2025

In-Hand Manipulation of Articulated Tools with Dexterous Robot Hands with Sim-to-Real Transfer

Reinforcement learning (RL) and sim-to-real transfer have advanced rigid-object manipulation. However, policies remain brittle for articulated mechanisms due to contact-rich dynamics that require both stable grasping and simultaneous free in-hand articulation. Furthermore, articulated objects and robot hands exhibit under-modeled joint phenomena such as friction, stiction, and backlash in real life that can increase the sim-to-real gap, and robot hands still fall short of idealized tactile sensing, both in terms of coverage, sensitivity, and specificity. In this paper, we present an original approach to learning dexterous in-hand manipulation of articulated tools that has reduced articulation and kinematic redundancy relative to the human hand. Our approach augments a simulation-trained base policy with a sensor-driven refinement learned from hardware demonstrations. This refinement conditions on proprioception and target articulation states while fusing whole-hand tactile and force-torque feedback with the policy's action intent through cross-attention. The resulting controller adapts online to instance-specific articulation properties, stabilizes contact interactions, and regulates internal forces under perturbations. We validate our method across diverse real-world tools, including scissors, pliers, minimally invasive surgical instruments, and staplers, demonstrating robust sim-to-real transfer, improved disturbance resilience, and generalization across structurally related articulated tools without precise physical modeling.

cs.RO↗

arXiv2025

A Computational Approach to Analyzing Language Change and Variation in the Constructed Language Toki Pona

This study explores language change and variation in Toki Pona, a constructed language with approximately 120 core words. Taking a computational and corpus-based approach, the study examines features including fluid word classes and transitivity in order to examine (1) changes in preferences of content words for different syntactic positions over time and (2) variation in usage across different corpora. The results suggest that sociolinguistic factors influence Toki Pona in the same way as natural languages, and that even constructed linguistic systems naturally evolve as communities use them.

cs.CL↗

arXiv2025

Scaling Gaussian Process Regression with Full Derivative Observations

We present a scalable Gaussian Process (GP) method called DSoftKI that can fit and predict full derivative observations. It extends SoftKI, a method that approximates a kernel via softmax interpolation, to the setting with derivatives. DSoftKI enhances SoftKI's interpolation scheme by replacing its global temperature vector with local temperature vectors associated with each interpolation point. This modification allows the model to encode local directional sensitivity, enabling the construction of a scalable approximate kernel, including its first and second-order derivatives, through interpolation. Moreover, the interpolation scheme eliminates the need for kernel derivatives, facilitating extensions such as Deep Kernel Learning (DKL). We evaluate DSoftKI on synthetic benchmarks, a toy n-body physics simulation, standard regression datasets with synthetic gradients, and high-dimensional molecular force field prediction (100-1000 dimensions). Our results demonstrate that DSoftKI is accurate and scales to larger datasets with full derivative observations than previously possible.

cs.LG↗

arXiv2024

High-Dimensional Gaussian Process Regression with Soft Kernel Interpolation

We introduce Soft Kernel Interpolation (SoftKI), a method that combines aspects of Structured Kernel Interpolation (SKI) and variational inducing point methods, to achieve scalable Gaussian Process (GP) regression on high-dimensional datasets. SoftKI approximates a kernel via softmax interpolation from a smaller number of interpolation points learned by optimizing a combination of the SoftKI marginal log-likelihood (MLL), and when needed, an approximate MLL for improved numerical stability. Consequently, it can overcome the dimensionality scaling challenges that SKI faces when interpolating from a dense and static lattice while retaining the flexibility of variational methods to adapt inducing points to the dataset. We demonstrate the effectiveness of SoftKI across various examples and show that it is competitive with other approximated GP methods when the data dimensionality is modest (around 10).

stat.ML↗

arXiv2023

Quantum Computing and Visualization: A Disruptive Technological Change Ahead

The focus of this Visualization Viewpoints article is to provide some background on Quantum Computing (QC), to explore ideas related to how visualization helps in understanding QC, and examine how QC might be useful for visualization with the growth and maturation of both technologies in the future. In a quickly evolving technology landscape, QC is emerging as a promising pathway to overcome the growth limits in classical computing. In some cases, QC platforms offer the potential to vastly outperform the familiar classical computer by solving problems more quickly or that may be intractable on any known classical platform. As further performance gains for classical computing platforms are limited by diminishing Moore's Law scaling, QC platforms might be viewed as a potential successor to the current field of exascale-class platforms. While present-day QC hardware platforms are still limited in scale, the field of quantum computing is robust and rapidly advancing in terms of hardware capabilities, software environments for developing quantum algorithms, and educational programs for training the next generation of scientists and engineers. After a brief introduction to QC concepts, the focus of this article is to explore the interplay between the fields of visualization and QC. First, visualization has played a role in QC by providing the means to show representations of the quantum state of single-qubits in superposition states and multiple-qubits in entangled states. Second, there are a number of ways in which the field of visual data exploration and analysis may potentially benefit from this disruptive new technology though there are challenges going forward.

quant-ph↗

arXiv2023

On Training Derivative-Constrained Neural Networks

We refer to the setting where the (partial) derivatives of a neural network's (NN's) predictions with respect to its inputs are used as additional training signal as a derivative-constrained (DC) NN. This situation is common in physics-informed settings in the natural sciences. We propose an integrated RELU (IReLU) activation function to improve training of DC NNs. We also investigate denormalization and label rescaling to help stabilize DC training. We evaluate our methods on physics-informed settings including quantum chemistry and Scientific Machine Learning (SciML) tasks. We demonstrate that existing architectures with IReLU activations combined with denormalization and label rescaling better incorporate training signal provided by derivative constraints.

cs.LG↗

arXiv2023

ExpeL: LLM Agents Are Experiential Learners

The recent surge in research interest in applying large language models (LLMs) to decision-making tasks has flourished by leveraging the extensive world knowledge embedded in LLMs. While there is a growing demand to tailor LLMs for custom decision-making tasks, finetuning them for specific tasks is resource-intensive and may diminish the model's generalization capabilities. Moreover, state-of-the-art language models like GPT-4 and Claude are primarily accessible through API calls, with their parametric weights remaining proprietary and unavailable to the public. This scenario emphasizes the growing need for new methodologies that allow learning from agent experiences without requiring parametric updates. To address these problems, we introduce the Experiential Learning (ExpeL) agent. Our agent autonomously gathers experiences and extracts knowledge using natural language from a collection of training tasks. At inference, the agent recalls its extracted insights and past experiences to make informed decisions. Our empirical results highlight the robust learning efficacy of the ExpeL agent, indicating a consistent enhancement in its performance as it accumulates experiences. We further explore the emerging capabilities and transfer learning potential of the ExpeL agent through qualitative observations and additional experiments.

cs.LG↗

arXiv2023

Push: Concurrent Probabilistic Programming for Bayesian Deep Learning

We introduce a library called Push that takes a probabilistic programming approach to Bayesian deep learning (BDL). This library enables concurrent execution of BDL inference algorithms on multi-GPU hardware for neural network (NN) models. To accomplish this, Push introduces an abstraction that represents an input NN as a particle. Push enables easy creation of particles so that an input NN can be replicated and particles can communicate asynchronously so that a variety of parameter updates can be expressed, including common BDL algorithms. Our hope is that Push lowers the barrier to experimenting with BDL by streamlining the scaling of particles across GPUs. We evaluate the scaling behavior of particles on single-node multi-GPU devices on vision and scientific machine learning (SciML) tasks.

cs.LG↗

arXiv2020

Elementary Logic in Linear Space

First-order logic is typically presented as the study of deduction in a setting with elementary quantification. In this paper, we take another vantage point and conceptualize first-order logic as a linear space that encodes "plausibility". Whereas a deductive perspective emphasizes how (i.e., process), a space perspective emphasizes where (i.e., location). We explore several consequences that a shift in perspective to "signals in space" has for first-order logic, including (1) a notion of proof based on orthogonal decomposition, (2) a method for assigning probabilities to sentences that reflects logical uncertainty, and (3) a "models as boundary" principle that relates the models of a theory to its "size".

cs.LO↗

arXiv2019

On Learning to Prove

In this paper, we consider the problem of learning a first-order theorem prover that uses a representation of beliefs in mathematical claims to construct proofs. The inspiration for doing so comes from the practices of human mathematicians where "plausible reasoning" is applied in addition to deductive reasoning to find proofs. Towards this end, we introduce a representation of beliefs that assigns probabilities to the exhaustive and mutually exclusive first-order possibilities found in Hintikka's theory of distributive normal forms. The representation supports Bayesian update, induces a distribution on statements that does not enforce that logically equivalent statements are assigned the same probability, and suggests an embedding of statements into an associated Hilbert space. We then examine conjecturing as model selection and an alternating-turn game of determining consistency. The game is amenable (in principle) to self-play training to learn beliefs and derive a prover that is complete when logical omniscience is attained and sound when beliefs are reasonable. The representation has super-exponential space requirements as a function of quantifier depth so the ideas in this paper should be taken as theoretical. We will comment on how abstractions can be used to control the space requirements at the cost of completeness.

cs.AI↗

arXiv2018

GamePad: A Learning Environment for Theorem Proving

In this paper, we introduce a system called GamePad that can be used to explore the application of machine learning methods to theorem proving in the Coq proof assistant. Interactive theorem provers such as Coq enable users to construct machine-checkable proofs in a step-by-step manner. Hence, they provide an opportunity to explore theorem proving with human supervision. We use GamePad to synthesize proofs for a simple algebraic rewrite problem and train baseline models for a formalization of the Feit-Thompson theorem. We address position evaluation (i.e., predict the number of proof steps left) and tactic prediction (i.e., predict the next proof step) tasks, which arise naturally in tactic-based theorem proving.

cs.LG↗

arXiv2018

An Application of Computable Distributions to the Semantics of Probabilistic Programs

In this chapter, we explore how (Type-2) computable distributions can be used to give both (algorithmic) sampling and distributional semantics to probabilistic programs with continuous distributions. Towards this end, we sketch an encoding of computable distributions in a fragment of Haskell and show how topological domains can be used to model the resulting PCF-like language. We also examine the implications that a (Type-2) computable semantics has for implementing conditioning. We hope to draw out the connection between an approach based on (Type-2) computability and ordinary programming throughout the chapter as well as highlight the relation with constructive mathematics (via realizability).

cs.PL↗

arXiv2014

Augur: a Modeling Language for Data-Parallel Probabilistic Inference

It is time-consuming and error-prone to implement inference procedures for each new probabilistic model. Probabilistic programming addresses this problem by allowing a user to specify the model and having a compiler automatically generate an inference procedure for it. For this approach to be practical, it is important to generate inference code that has reasonable performance. In this paper, we present a probabilistic programming language and compiler for Bayesian networks designed to make effective use of data-parallel architectures such as GPUs. Our language is fully integrated within the Scala programming language and benefits from tools such as IDE support, type-checking, and code completion. We show that the compiler can generate data-parallel inference code scalable to thousands of GPU cores by making use of the conditional independence relationships in the Bayesian network.

stat.ML↗