arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Author

Jacob Jackson

20 papers · 2003–2026

2003: 12004: 02005: 02006: 02007: 02008: 02009: 02010: 12011: 12012: 22013: 12014: 12015: 02016: 22017: 32018: 02019: 22020: 12021: 22022: 02023: 12024: 02025: 12026: 1
2003peak 3 in 20172026

Frequent co-authors

Papers

showing 1–20 of 20 RSS

Scaling Laws for Autoregressive Generative Modeling

Tom Henighan, Jared Kaplan, Mor Katz +16 more

cs.LG · · cited 664× · arXiv ↗ · more like this ·

We identify empirical scaling laws for the cross-entropy loss in four domains: generative image modeling, video modeling, multimodal image$\leftrightarrow$text models, and mathematical problem solving. In all cases autoregressive Transformers smoothly improve in performance as model size and compute budgets increase, following a power-law plus constant scaling law. The optimal model size also depends on the compute budget through a power-law, with exponents that are nearly universal across all data domains. The cross-entropy loss has an information theoretic interpretation as $S($True$) + D_{\mathrm{KL}}($True$||$Model$)$, and the empirical scaling laws suggest a prediction for both the true data distribution's entropy and the KL divergence between the true and model distributions. With this interpretation, billion-parameter Transformers are nearly perfect models of the YFCC100M image…

College Readiness and College Completion at Broad Access Four-Year Institutions

Jacob Jackson, Michal Kurlaender

Education · American Behavioral Scientist · · cited 49× · DOI ↗ · more like this ·

Many students arrive at college unprepared to do college-level work, facing a host of college readiness assessments and developmental education coursework. In this article, we examine how a student’s readiness to take college classes upon entry to the university is related to four longer term postsecondary outcomes. We utilize a unique longitudinal dataset of the census of first-time freshmen at the nation’s largest public 4-year state postsecondary system to describe differences by campus and across college majors. In addition we explore the use of one measure that the system does not currently use to determine readiness—high school grade point average—and its relationship with both the existing readiness measure and postsecondary outcomes. Our results indicate that grade point average can be a useful predictor of college success above and beyond the readiness measure.

Scaling of urban income inequality in the USA

Elisa Heinrich Mora, Cate Heine, Jacob J. Jackson +3 more

Sociology and Political Science · Journal of The Royal Society Interface · · cited 41× · DOI ↗ · more like this ·

Urban scaling analysis, the study of how aggregated urban features vary with the population of an urban area, provides a promising framework for discovering commonalities across cities and uncovering dynamics shared by cities across time and space. Here, we use the urban scaling framework to study an important, but under-explored feature in this community-income inequality. We propose a new method to study the scaling of income distributions by analysing total income scaling in population percentiles. We show that income in the least wealthy decile (10%) scales close to linearly with city population, while income in the most wealthy decile scale with a significantly superlinear exponent. In contrast to the superlinear scaling of total income with city population, this decile scaling illustrates that the benefits of larger cities are increasingly unequally distributed. For the poorest…

Semi-Supervised Learning by Label Gradient Alignment

Jacob Jackson, John Schulman

cs.LG · · cited 20× · arXiv ↗ · more like this ·

We present label gradient alignment, a novel algorithm for semi-supervised learning which imputes labels for the unlabeled data and trains on the imputed labels. We define a semantically meaningful distance metric on the input space by mapping a point (x, y) to the gradient of the model at (x, y). We then formulate an optimization problem whose objective is to minimize the distance between the labeled and the unlabeled data in this space, and we solve it by gradient descent on the imputed labels. We evaluate label gradient alignment using the standardized architecture introduced by Oliver et al. (2018) and demonstrate state-of-the-art accuracy in semi-supervised CIFAR-10 classification.

Does an Early College Readiness Signal Discourage College Application and Enrollment?

Jacob Jackson

Education · Journal of Research on Educational Effectiveness · · cited 12× · DOI ↗ · more like this ·

States are currently seeking ways to improve alignment between K–12 and postsecondary education and better prepare students for postsecondary schooling. Some states have begun implementing policies that inform high school students that they are not ready for college-level courses well before they graduate, in order to give both notice and time to prepare. However, if the early information from these tests suggests that a student may require remedial coursework upon entry to college, the student may be discouraged from attending college. This article focuses on one state's effort to increase students’ information about college readiness. Regression discontinuity design is used to determine whether a student receiving a signal that he/she needs remediation lowers the probability of that student applying to or enrolling in participating universities. Results indicate that the policy has…

An Empirical Study on Hyperparameters and their Interdependence for RL Generalization

Xingyou Song, Yilun Du, Jacob Jackson

cs.LG · · cited 9× · arXiv ↗ · more like this ·

Recent results in Reinforcement Learning (RL) have shown that agents with limited training environments are susceptible to a large amount of overfitting across many domains. A key challenge for RL generalization is to quantitatively explain the effects of changing parameters on testing performance. Such parameters include architecture, regularization, and RL-dependent variables such as discount factor and action stochasticity. We provide empirical results that show complex and interdependent relationships between hyperparameters and generalization. We further show that several empirical metrics such as gradient cosine similarity and trajectory-dependent metrics serve to provide intuition towards these results.

K–12 Postsecondary Alignment and School Accountability: Investigating High School Responses to California’s Early Assessment Program

Jacob Jackson, Michal Kurlaender

Education · American Journal of Education · · cited 7× · DOI ↗ · more like this ·

State K–12 assessments may soon include measures for college readiness, as California’s already do. We seek to understand how California’s Early Assessment Program (EAP, designed to assess high school juniors’ college readiness in English and math) may have influenced overall school-level college readiness and state accountability outcomes. Using administrative data, we first examine variation in EAP participation across California’s public high schools and what accounts for the variation in the early years of the program. Second, we ask whether different levels of EAP participation have influenced school outcomes. Finally, we ask whether the EAP has led to changes in schoolwide accountability measures by increasing eleventh-grade students’ testing performance. We find that schools with higher participation in these aligned assessments witnessed small improvements on the state’s primary…

Learning nonlinearly separable mod k addition problem using a single multi-valued neuron with a periodic activation function

Igor Aizenberg, Matthew Caudill, Jacob Jackson +1 more

Artificial Intelligence · · cited 6× · DOI ↗ · more like this ·

In this paper, we further develop a complex-valued neuron paradigm. It is shown how a single multi-valued neuron with a periodic activation function may learn multiple-valued nonlinearly separable problems. One of the classical nonlinearly separable problems - mod k addition of n variables is considered in detail. It is shown that to be able to learn this problem using a single multi-valued neuron, it is necessary to use a periodic activation function and a learning algorithm based on the error-correction learning rule and adapted to this activation function.

Investigating Middle School Determinants of High School Achievement and Graduation in Three California School Districts

Michal Kurlaender, Jacob Jackson

Education · California Journal of Politics and Policy · · cited 5× · DOI ↗ · more like this ·

This paper explores early predictors of high school graduation and success. Employing 7th grade cohorts from three large California school districts (San Francisco, Fresno, and Long Beach), we investigate the role of several key middle school academic performance measures in identifying students' 11th grade academic performance, passing the California High School Exit Examination on the first attempt, and diploma receipt. We find that standardized assessments, timing of algebra, and course failures in middle school provide useful indicators of students' high school academic success. Our aim is not to identify any causal mechanism by which middle school achievement leads to high school success or failure, but rather to describe important associations that may aid policymakers and school leaders to develop strategies early in students' educational pursuit of the high school diploma.

Intelligent edge enhancement using multilayer neural network based on multi-valued neurons

Igor Aizenberg, Shane Alexander, Jacob Jackson +3 more

Artificial Intelligence · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · · cited 3× · DOI ↗ · more like this ·

In this paper, we solve the edge enhancement problem using an intelligent approach. We use a multilayer neural network based on multi-valued neurons (MLMVN) as an intelligent edge enhancer. The problem of neural edge enhancement using a classical multilayer feedforward neural network (MLF) was already considered by some authors. Since MLMVN significantly outperforms MLF in terms of learning speed, number of parameters employed, and generalization capability, it is very attractive to apply it for solving the edge enhancement problem. The main result which is presented in the paper, is the proven ability of MLMVN to enhance edges corresponding to a certain edge detection operator. Moreover, it is possible to enhance edges on noisy images ignoring a noisy texture. It is shown that to learn any edge detection operator using MLMVN, only a single image is required for learning purposes. The…

<title>Wavelet-based steganalysis using a computational immune system approach</title>

Jacob Jackson, Gregg H. Gunsch, Roger L. Claypoole +1 more

Computer Vision and Pattern Recognition · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · · cited 3× · DOI ↗ · more like this ·

Current signature-based approaches for detecting the presence of hidden messages in digital files - the initial step in steganalysis - rely on discerning "footprints" of steganographic tools. Of greater recent concern is detecting the use of novel proprietary or "home-grown" tools for which no signature has been established. This research focuses on detecting anomalies in seemingly innocuous images, applying a genetic algorithm within a computational immune system to leverage powerful image processing through wavelet analysis. The sensors developed with this new, synergistic system demonstrated a surprising level of capability to detect the use of steganographic tools for which the system had no previous exposure.

Higher Education in California: Expanding College Access

Jacob Jackson, Sarah Bohn, Hans Johnson

Education · VTechWorks (Virginia Tech) · · cited 2× · OpenAlex ↗ · more like this ·

More California high school graduates are academically ready for college than ever before. More are applying to and enrolling in college, and both the University of California (UC) and the California State University (CSU) are expanding access: for example, UC has pledged to enroll 10,000 more freshmen and transfer students between 2016 and 2018. But many qualified applicants are being turned away, and this is a source of concern. A 2016 PPIC Statewide Survey found that more than three-fourths of Californians are concerned about access to the UC system. Expanding access to college benefits both individuals and the state as a whole. The economic returns to a postsecondary degree are at their highest level in decades, even as more Californians are attending college; and workers with postsecondary degrees will continue to play a crucial role in the state’s economic growth. Expanding access…

An Isometric Stochastic Optimizer

Jacob Jackson

cs.LG · · cited 1× · arXiv ↗ · more like this ·

The Adam optimizer is the standard choice in deep learning applications. I propose a simple explanation of Adam's success: it makes each parameter's step size independent of the norms of the other parameters. Based on this principle I derive Iso, a new optimizer which makes the norm of a parameter's update invariant to the application of any linear transformation to its inputs and outputs. I develop a variant of Iso called IsoAdam that allows optimal hyperparameters to be transferred from Adam, and demonstrate that IsoAdam obtains a speedup over Adam when training a small Transformer.

Remedial Education in California’s Colleges and Universities

Olga Rodríguez, Jacob Jackson, Marisol Cuellar Mejia

Education · VTechWorks (Virginia Tech) · · cited 1× · OpenAlex ↗ · more like this ·

Remedial courses, also known as developmental or basic skills education, generally cover high school material and are aimed at students deemed unprepared for college-level work. At CSU, Latino and African American students are about twice as likely to be placed in a remedial class as their white and Asian American counterparts. At CCC, all groups enroll in remediation at high rates; however, Latino and African American students are more likely to enroll in remedial courses than their white and Asian American peers. What is more, Latinos and African Americans are also more likely to enroll in lower levels of remediation. Similar patterns occur with low-income students and their higher-income peers at both CSU and CCC.

IMPACT OF SOCIAL STUDIES CURRICULUM IN PROMOTING SOCIAL INTEGRATION AMONG PRIMARY SCHOOLS PUPILS IN GOMBI EDUCATIONAL ZONE, ADAMAWA STATE, NIGERIA

SUNDAY TUNFANG, JACOB JACKSON

Education · International Journal of Library Science and Educational Research · · cited 0× · DOI ↗ · more like this ·

This study investigated “impact of social studies curriculum in promoting social integration among primary schools pupils in Gombi Educational Zone, Adamawa State, Nigeria. Two (2) research objectives, questions and hypotheses were formulated to guide the study. The hypotheses were formulated and tested at P< 0.05 level of significance. Descriptive survey design was adopted for the study. The population of the study was made up of 98 from 8 selected private and public primary schools in Gombi educational zone. The sample size was 98 made up of 57 male and 41 female social studies teachers. The study used census sampling technique for whole social studies teachers. “Impact of Social Integration among Primary School Pupils Questionnaire (SIPAPSQ)" as an instrument for data collection. The reliability of the instrument was obtained through the pilot test that was conducted for the…

Scaling of Urban Income Inequality in the United States

Elisa Heinrich Mora, Jacob J. Jackson, Cate Heine +3 more

physics.soc-ph · · cited 0× · arXiv ↗ · more like this ·

Urban scaling analysis, the study of how aggregated urban features vary with the population of an urban area, provides a promising framework for discovering commonalities across cities and uncovering dynamics shared by cities across time and space. Here, we use the urban scaling framework to study an important, but under-explored feature in this community - income inequality. We propose a new method to study the scaling of income distributions by analyzing total income scaling in population percentiles. We show that income in the least wealthy decile (10%) scales close to linearly with city population, while income in the most wealthy decile scale with a significantly superlinear exponent. In contrast to the superlinear scaling of total income with city population, this decile scaling illustrates that the benefits of larger cities are increasingly unequally distributed. For the poorest…

Technology: Key to Student Success or Another Distraction?

Jacob Jackson

Education · Digital Commons - Andrews University (Andrews University) · · cited 0× · OpenAlex ↗ · more like this ·

In this digital age, many educational institutions have pushed for a deeper integration of technology into the classroom. But is new and flashy always better? To be sure, technology is useful for improving a student's computer literacy, but at what point do such attempts become a redundancy, and when do they hinder a student's ability to focus on tasks at hand? This presentation aims to address these questions by critically analyzing the trend of increasingly technological classrooms and technology’s effect on a student productivity and achievement.

Summary Notes

Michal Kurlaender, Jacob Jackson, Jessica S. Howell

Education · · cited 0× · OpenAlex ↗ · more like this ·

higher education under California’s Early Assessment Program (EAP) may improve measures of school performance. • Proficiency levels on the state standardized test, school accountability measures, and college application rates improved at all high schools following the introduction of the EAP. Gains in these measures of school performance were strongest at those high schools with the largest proportion of juniors participating in the program. • Results suggest that college readiness standards that are measured appropriately by well-aligned assessments may improve measures of high school accountability within a state. The Common Core State Standards movement has driven policy discussion on the need to better align K–12 systems with higher education within and across states to ensure a more seamless transition for young adults between high school and college. States have adopted different…

Composer 2 Technical Report

Cursor Research, :, Aaron Chan +53 more

cs.SE · · arXiv ↗ · more like this ·

Composer 2 is a specialized model designed for agentic software engineering. The model demonstrates strong long-term planning and coding intelligence while maintaining the ability to efficiently solve problems for interactive use. The model is trained in two phases: first, continued pretraining to improve the model's knowledge and latent coding ability, followed by large-scale reinforcement learning to improve end-to-end coding performance through stronger reasoning, accurate multi-step execution, and coherence on long-horizon realistic coding problems. We develop infrastructure to support training in the same Cursor harness that is used by the deployed model, with equivalent tools and structure, and use environments that match real problems closely. To measure the ability of the model on increasingly difficult tasks, we introduce a benchmark derived from real software engineering…