Skip to Content
logologo
AI Incident Database
Open TwitterOpen RSS FeedOpen FacebookOpen LinkedInOpen GitHub
Open Menu
Descubrir
Enviar
  • Bienvenido a la AIID
  • Descubrir Incidentes
  • Vista espacial
  • Vista Tabular
  • Vista de lista
  • Entidades
  • Taxonomías
  • Enviar Informes de Incidentes
  • Ranking de Reportadores
  • Blog
  • Resumen de noticias de IA
  • Control de Riesgos
  • Incidente aleatorio
  • Registrarse
Colapsar
Descubrir
Enviar
  • Bienvenido a la AIID
  • Descubrir Incidentes
  • Vista espacial
  • Vista Tabular
  • Vista de lista
  • Entidades
  • Taxonomías
  • Enviar Informes de Incidentes
  • Ranking de Reportadores
  • Blog
  • Resumen de noticias de IA
  • Control de Riesgos
  • Incidente aleatorio
  • Registrarse
Colapsar
Traducido por IA

Incidente 65: Funciones de recompensa del aprendizaje por refuerzo en los videojuegos

Traducido por IA
Descripción:
Traducido por IA
OpenAI publicó un post sobre sus hallazgos al utilizar Universe, un software para medir y entrenar agentes de IA para realizar experimentos de aprendizaje de refuerzo, mostrando que el agente de IA no actuó de la manera prevista para completar un videojuego.

Herramientas

Nuevo InformeNuevo InformeNueva RespuestaNueva RespuestaDescubrirDescubrirVer HistorialVer Historial

Entidades

Ver todas las entidades
Presunto: un sistema de IA desarrollado e implementado por OpenAI, perjudicó a OpenAI.

Estadísticas de incidentes

ID
65
Cantidad de informes
1
Fecha del Incidente
2016-12-22
Editores
Sean McGregor
Applied Taxonomies
CSETv0, CSETv1, GMF, MIT

Clasificaciones de la Taxonomía CSETv1

Detalles de la Taxonomía

Incident Number

The number of the incident in the AI Incident Database.
 

65

Special Interest Intangible Harm

An assessment of whether a special interest intangible harm occurred. This assessment does not consider the context of the intangible harm, if an AI was involved, or if there is characterizable class or subgroup of harmed entities. It is also not assessing if an intangible harm occurred. It is only asking if a special interest intangible harm occurred.
 

no

Date of Incident Year

The year in which the incident occurred. If there are multiple harms or occurrences of the incident, list the earliest. If a precise date is unavailable, but the available sources provide a basis for estimating the year, estimate. Otherwise, leave blank. Enter in the format of YYYY
 

2016

Date of Incident Month

The month in which the incident occurred. If there are multiple harms or occurrences of the incident, list the earliest. If a precise date is unavailable, but the available sources provide a basis for estimating the month, estimate. Otherwise, leave blank. Enter in the format of MM
 

12

Date of Incident Day

The day on which the incident occurred. If a precise date is unavailable, leave blank. Enter in the format of DD
 

21

Estimated Date

“Yes” if the data was estimated. “No” otherwise.
 

Yes

Clasificaciones de la Taxonomía CSETv0

Detalles de la Taxonomía

Problem Nature

Indicates which, if any, of the following types of AI failure describe the incident: "Specification," i.e. the system's behavior did not align with the true intentions of its designer, operator, etc; "Robustness," i.e. the system operated unsafely because of features or changes in its environment, or in the inputs the system received; "Assurance," i.e. the system could not be adequately monitored or controlled during operation.
 

Specification

Physical System

Where relevant, indicates whether the AI system(s) was embedded into or tightly associated with specific types of hardware.
 

Software only

Level of Autonomy

The degree to which the AI system(s) functions independently from human intervention. "High" means there is no human involved in the system action execution; "Medium" means the system generates a decision and a human oversees the resulting action; "low" means the system generates decision-support output and a human makes a decision and executes an action.
 

Unclear/unknown

Nature of End User

"Expert" if users with special training or technical expertise were the ones meant to benefit from the AI system(s)’ operation; "Amateur" if the AI systems were primarily meant to benefit the general public or untrained users.
 

Expert

Public Sector Deployment

"Yes" if the AI system(s) involved in the accident were being used by the public sector or for the administration of public goods (for example, public transportation). "No" if the system(s) were being used in the private sector or for commercial purposes (for example, a ride-sharing company), on the other.
 

No

Data Inputs

A brief description of the data that the AI system(s) used or were trained on.
 

Universe software training

Clasificaciones de la Taxonomía MIT

Machine-Classified
Detalles de la Taxonomía

Risk Subdomain

A further 23 subdomains create an accessible and understandable classification of hazards and harms associated with AI
 

7.1. AI pursuing its own goals in conflict with human goals or values

Risk Domain

The Domain Taxonomy of AI Risks classifies risks into seven AI risk domains: (1) Discrimination & toxicity, (2) Privacy & security, (3) Misinformation, (4) Malicious actors & misuse, (5) Human-computer interaction, (6) Socioeconomic & environmental harms, and (7) AI system safety, failures & limitations.
 
  1. AI system safety, failures, and limitations

Entity

Which, if any, entity is presented as the main cause of the risk
 

AI

Timing

The stage in the AI lifecycle at which the risk is presented as occurring
 

Post-deployment

Intent

Whether the risk is presented as occurring as an expected or unexpected outcome from pursuing a goal
 

Unintentional

Informes del Incidente

Cronología de Informes

+1
Funciones de recompensa defectuosas en la naturaleza
Funciones de recompensa defectuosas en la naturaleza

Funciones de recompensa defectuosas en la naturaleza

blog.openai.com

Funciones de recompensa defectuosas en la naturaleza
blog.openai.com · 2016
Traducido por IA

En OpenAI, recientemente comenzamos a usar Universe, nuestro software para medir y entrenar agentes de IA, para realizar nuevos experimentos de RL. A veces, estos experimentos ilustran algunos de los problemas con RL tal como se practica ac…

Variantes

Una "Variante" es un incidente de IA similar a un caso conocido—tiene los mismos causantes, daños y sistema de IA. En lugar de enumerarlo por separado, lo agrupamos bajo el primer incidente informado. A diferencia de otros incidentes, las variantes no necesitan haber sido informadas fuera de la AIID. Obtenga más información del trabajo de investigación.
¿Has visto algo similar?

Incidentes Similares

Por similitud de texto

Did our AI mess up? Flag the unrelated incidents

Biased Sentiment Analysis

Biased Sentiment Analysis

Oct 2017 · 7 informes
Gender Biases in Google Translate

Gender Biases in Google Translate

Apr 2017 · 10 informes
Tesla Autopilot’s Lane Recognition Allegedly Vulnerable to Adversarial Attacks

Tesla Autopilot’s Lane Recognition Allegedly Vulnerable to Adversarial Attacks

Mar 2019 · 1 informe
Incidente AnteriorSiguiente Incidente

Incidentes Similares

Por similitud de texto

Did our AI mess up? Flag the unrelated incidents

Biased Sentiment Analysis

Biased Sentiment Analysis

Oct 2017 · 7 informes
Gender Biases in Google Translate

Gender Biases in Google Translate

Apr 2017 · 10 informes
Tesla Autopilot’s Lane Recognition Allegedly Vulnerable to Adversarial Attacks

Tesla Autopilot’s Lane Recognition Allegedly Vulnerable to Adversarial Attacks

Mar 2019 · 1 informe

Investigación

  • Definición de un “Incidente de IA”
  • Definición de una “Respuesta a incidentes de IA”
  • Hoja de ruta de la base de datos
  • Trabajo relacionado
  • Descargar Base de Datos Completa

Proyecto y Comunidad

  • Acerca de
  • Contactar y Seguir
  • Aplicaciones y resúmenes
  • Guía del editor

Incidencias

  • Todos los incidentes en forma de lista
  • Incidentes marcados
  • Cola de envío
  • Vista de clasificaciones
  • Taxonomías

2024 - AI Incident Database

  • Condiciones de uso
  • Política de privacidad
  • Open twitterOpen githubOpen rssOpen facebookOpen linkedin
  • 69ff178