arXiv subjects

Fernando Delgado

Publications and source records attributed to Fernando Delgado.

10 recordsLinked to original sources

arXiv2024

Machine Unlearning Doesn't Do What You Think: Lessons for Generative AI Policy and Research

"Machine unlearning" is a popular proposed solution for mitigating the existence of content in an AI model that is problematic for legal or moral reasons, including privacy, copyright, safety, and more. For example, unlearning is often invoked as a solution for removing the effects of specific information from a generative-AI model's parameters, e.g., a particular individual's personal data or the inclusion of copyrighted content in the model's training data. Unlearning is also proposed as a way to prevent a model from generating targeted types of information in its outputs, e.g., generations that closely resemble a particular individual's data or reflect the concept of "Spiderman." Both of these goals--the targeted removal of information from a model and the targeted suppression of information from a model's outputs--present various technical and substantive challenges. We provide a framework for ML researchers and policymakers to think rigorously about these challenges, identifying several mismatches between the goals of unlearning and feasible implementations. These mismatches explain why unlearning is not a general-purpose solution for circumscribing generative-AI model behavior in service of broader positive impact.

cs.LG↗

arXiv2023

The Participatory Turn in AI Design: Theoretical Foundations and the Current State of Practice

Despite the growing consensus that stakeholders affected by AI systems should participate in their design, enormous variation and implicit disagreements exist among current approaches. For researchers and practitioners who are interested in taking a participatory approach to AI design and development, it remains challenging to assess the extent to which any participatory approach grants substantive agency to stakeholders. This article thus aims to ground what we dub the "participatory turn" in AI design by synthesizing existing theoretical literature on participation and through empirical investigation and critique of its current practices. Specifically, we derive a conceptual framework through synthesis of literature across technology design, political theory, and the social sciences that researchers and practitioners can leverage to evaluate approaches to participation in AI design. Additionally, we articulate empirical findings concerning the current state of participatory practice in AI design based on an analysis of recently published research and semi-structured interviews with 12 AI researchers and practitioners. We use these empirical findings to understand the current state of participatory practice and subsequently provide guidance to better align participatory goals and methods in a way that accounts for practical constraints.

cs.HC↗

arXiv2022

A theoretical review on the single-impurity electron spin resonance on surfaces

The development of electron spin resonance (ESR) combined with scanning tunneling spectroscopy (STM) is undoubtedly one of the main experimental breakthroughs in surface science of the last decade thanks to joining the extraordinarily high energy resolution of ESR (nano-eV scale) with the single-atom spatial resolution of STM (sub-\AA ngstr\"om scale). While the experimental results have \fer{significantly} grown with the number of groups that have succeeded in implementing the technique, the physical mechanism behind it is still unclear, with several different mechanisms proposed to explain it. Here, we start by revising the main characteristics of the experimental setups and observed features. Then, we review the main theoretical proposals, with both their strengths and weaknesses. One of our conclusions is that many of the proposed mechanisms share the same basic principles, \fer{the time-dependent electric field at the STM junction is modulating the coupling of the spin-polarized transport electrons with the local spin.} This explains why these mechanims are essentially equivalent in a broad picture. We analyze the subtle differences between some of them and how they compare with the different experimental observations.

cond-mat.str-el↗

arXiv2022

An Uncommon Task: Participatory Design in Legal AI

Despite growing calls for participation in AI design, there are to date few empirical studies of what these processes look like and how they can be structured for meaningful engagement with domain experts. In this paper, we examine a notable yet understudied AI design process in the legal domain that took place over a decade ago, the impact of which still informs legal automation efforts today. Specifically, we examine the design and evaluation activities that took place from 2006 to 2011 within the TeXT Retrieval Conference's (TREC) Legal Track, a computational research venue hosted by the National Institute of Standards and Technologies. The Legal Track of TREC is notable in the history of AI research and practice because it relied on a range of participatory approaches to facilitate the design and evaluation of new computational techniques--in this case, for automating attorney document review for civil litigation matters. Drawing on archival research and interviews with coordinators of the Legal Track of TREC, our analysis reveals how an interactive simulation methodology allowed computer scientists and lawyers to become co-designers and helped bridge the chasm between computational research and real-world, high-stakes litigation practice. In analyzing this case from the recent past, our aim is to empirically ground contemporary critiques of AI development and evaluation and the calls for greater participation as a means to address them.

cs.CY↗

arXiv2021

Stakeholder Participation in AI: Beyond "Add Diverse Stakeholders and Stir"

There is a growing consensus in HCI and AI research that the design of AI systems needs to engage and empower stakeholders who will be affected by AI. However, the manner in which stakeholders should participate in AI design is unclear. This workshop paper aims to ground what we dub a 'participatory turn' in AI design by synthesizing existing literature on participation and through empirical analysis of its current practices via a survey of recent published research and a dozen semi-structured interviews with AI researchers and practitioners. Based on our literature synthesis and empirical research, this paper presents a conceptual framework for analyzing participatory approaches to AI design and articulates a set of empirical findings that in ensemble detail out the contemporary landscape of participatory practice in AI design. These findings can help bootstrap a more principled discussion on how PD of AI should move forward across AI, HCI, and other research communities.

cs.AI↗

arXiv2021

A study of all-electric electron spin resonance using Floquet quantum master equations

We present a theoretical framework to describe experiments directed to controlling single-atom spin dynamics by electrical means using a scanning tunneling microscope. We propose a simple model consisting of a quantum impurity connected to electrodes where an electrical time-dependent bias is applied. We solve the problem in the limit of weak coupling between the impurity and the electrodes by means of a quantum master equation that is derived by the non-equilibrium Green's function formalism. We show results in two cases. The first case is just a single atomic orbital subjected to a time-dependent electric field, and the second case consists of a single atomic orbital coupled to a second spin-1/2. The first case reproduces the main experimental features Ti atoms on MgO/Ag (100) while the second one directly addresses the experiments on two Ti atoms. These calculations permit us to explore the effect of different parameters on the driving of the atomic spins as well as to reproduce experimental fingerprints.

cond-mat.mes-hall↗

arXiv2021

Spin Resonance Amplitude and Frequency of a Single Atom on a Surface in a Vector Magnetic Field

We used electron spin resonance (ESR) combined with scanning tunneling microscopy (STM) to measure hydrogenated Ti (spin-1/2) atoms at low-symmetry binding sites on MgO in vector magnetic fields. We found strongly anisotropic g-values in all three spatial directions. Interestingly, the amplitude and lineshape of the ESR signals are also strongly dependent on the angle of the field. We conclude that the Ti spin is aligned along the magnetic field, while the tip spin follows its strong magnetic anisotropy. Our results show the interplay between the tip and surface spins in determining the ESR signals and highlight the precision of ESR-STM to identify the single atom's spin states.

cond-mat.mes-hall↗

arXiv2019

Large effect of the metal substrate on the magnetic anisotropy of Co on hexagonal Boron Nitride

We combine x-ray absorption spectroscopy (XAS), x-ray magnetic circular dichroism (XMCD) and x-ray magnetic linear dichroism (XMLD) data with first principles density functional theory (DFT) calculations and a multiorbital many body Hamiltonian approach to understand the electronic and magnetic properties of Co atoms adsorbed on h-BN/Ru(0001) and h-BN/Ir(111). The XAS line shape reveals, for both substrates, an electronic configuration close to $3d^8$, corresponding to a spin $S = 1$. Magnetic field dependent XMCD data show large (14 meV) out-of-plane anisotropy on h-BN/Ru(0001), while it is almost isotropic (tens of $\mu$eV) on h-BN/Ir(111). XMLD data together with both DFT calculations and the results of the multiorbital Hubbard model suggest that the dissimilar magnetic anisotropy originates from different Co adsorption sites, namely atop N on h-BN/Ru(0001) and 6-fold hollow on h-BN/Ir(111).

cond-mat.mes-hall↗

arXiv2014

Derivation of the spin Hamiltonians for Fe in MgO

A method to calculate the effective spin Hamiltonian for a transition metal impurity in a non- magnetic insulating host is presented and applied to the paradigmatic case of Fe in MgO. In a first step we calculate the electronic structure employing standard density functional theory (DFT), based on generalized-gradient approximation (GGA), using plane waves as a basis set. The corresponding basis of atomic-like maximally localized Wannier functions is derived and used to represent the DFT Hamiltonian, resulting in a tight-binding model for the atomic orbitals of the magnetic impurity. The third step is to solve, by exact numerical diagonalization, the N electron problem in the open shell of the magnetic atom, including both effect of spin-orbit and Coulomb repulsion. Finally, the low energy sector of this multi-electron Hamiltonian is mapped into effective spin models that, in addition to the spin matrices S, can also include the orbital angular momentum L when appropriate. We successfully apply the method to Fe in MgO, considering both, the undistorted and Jahn-Teller (JT) distorted cases. Implications for the influence of Fe impurities on the performance of magnetic tunnel junctions based on MgO are discussed.

cond-mat.mes-hall↗

arXiv2014

Control of single-spin magnetic anisotropy by exchange coupling

The properties of quantum systems interacting with their environment, commonly called open quantum systems, can be strongly affected by this interaction. While this can lead to unwanted consequences, such as causing decoherence in qubits used for quantum computation, it can also be exploited as a probe of the environment. For example, magnetic resonance imaging is based on the dependence of the spin relaxation times of protons in water molecules in a host's tissue. Here we show that the excitation energy of a single spin, which is determined by magnetocrystalline anisotropy and controls its stability and suitability for use in magnetic data storage devices, can be modified by varying the exchange coupling of the spin to a nearby conductive electrode. Using scanning tunnelling microscopy and spectroscopy, we observe variations up to a factor of two of the spin excitation energies of individual atoms as the strength of the spin's coupling to the surrounding electronic bath changes. These observations, combined with calculations, show that exchange coupling can strongly modify the magnetic anisotropy. This system is thus one of the few open quantum systems in which the energy levels, and not just the excited-state lifetimes, can be controllably renormalized. Furthermore, we demonstrate that the magnetocrystalline anisotropy, a property normally determined by the local structure around a spin, can be electronically tuned. These effects may play a significant role in the development of spintronic devices5 in which an individual magnetic atom or molecule is coupled to conducting leads.

cond-mat.mes-hall↗