Skip to main content
Skip to content
Case File
kaggle-ho-013056House Oversight

Academic discussion of counterfactual models for assessing intelligent agents

Academic discussion of counterfactual models for assessing intelligent agents The passage is a theoretical exposition on AI evaluation methods with no mention of political figures, financial transactions, or misconduct. It offers no actionable investigative leads. Key insights: Describes a possible‑worlds semantics for pragmatic intelligence.; Introduces inference agents to model real‑world agents.; Defines naturalistic context with constant goals and rewards.

Date
Unknown
Source
House Oversight
Reference
kaggle-ho-013056
Pages
1
Persons
2
Integrity
No Hash Available

Summary

Academic discussion of counterfactual models for assessing intelligent agents The passage is a theoretical exposition on AI evaluation methods with no mention of political figures, financial transactions, or misconduct. It offers no actionable investigative leads. Key insights: Describes a possible‑worlds semantics for pragmatic intelligence.; Introduces inference agents to model real‑world agents.; Defines naturalistic context with constant goals and rewards.

Tags

kagglehouse-oversightai-theoryintelligence-measurementcounterfactual-modeling

Ask AI About This Document

0Share
PostReddit
Review This Document

Extracted Text (OCR)

EFTA Disclosure
Text extracted via OCR from the original document. May contain errors from the scanning process.
140 7 A Formal Model of Intelligent Agents We suggest to view the definitions of pragmatic and efficient pragmatic general intelligence in terms of a “possible worlds” semantics — i.e. to view them as asking, counterfactually, how an agent would perform, hypothetically, on a series of tests (the tests being goals, defined in relation to environments and reward signals). Real-world intelligent agents don’t normally operate in terms of explicit goals and rewards; these are abstractions that we use to think about intelligent agents. However, this is no objection to characterizing various sorts of intelligence in terms of counterfactuals like: how would system S operate if it were trying to achieve this or that goal, in this or that environment, in order to seek reward? We can characterize various sorts of intelligence in terms of how it can be inferred an agent would perform on certain tests, even though the agent’s real life does not consist of taking these tests. This conceptual approach may seem a bit artificial but we don’t currently see a better alternative, if one wishes to quantitatively gauge intelligence (which is, in a sense, an “artificial” thing to do in the first place). Given a real-world agent X and a mandate to assess its intelligence, the obvious alternative to looking at possible worlds in the manner of the above definitions, is just looking directly at the properties of the things X has achieved in the real world during its lifespan. But this isn’t an easy solution, because it doesn’t disambiguate which aspects of X’s achievements were due to its own actions versus due to the rest of the world that X was interacting with when it made its achievements. To distinguish the amount of achievement that X “caused” via its own actions requires a model of causality, which is a complex can of worms in itself; and, critically, the standard models of causality also involve counterfactuals (asking “what would have been achieved in this situation if the agent X hadn’t been there”, etc.) [MW07]. Regardless of the particulars, it seems impossible to avoid counterfactual realities in assessing intelligence. The approach we suggest — given a real-world agent X with a history of actions in a particular world, and a mandate to assess its intelligence — is to introduce an additional player, an inference agent 6, into the picture. The agent 7 modeled above is then viewed as 7x: the model of X that 6 constructs, in order to explore X’s inferred behaviors in various counterfactual environments. In the test situations embodied in the definitions of pragmatic and efficient pragmatic general intelligence, the environment gives 7x rewards, based on specifically configured goals. In X’s real life, the relation between goals, rewards and actions will generally be significantly subtler and perhaps quite different. We model the real world similarly to the “fantasy world” of the previous section, but with the omission of goals and rewards. We define a naturalistic context as one in which all goals and rewards are constant, i.e. g; = 99 and r; = 7p for all 7. This is just a mathematical convention for stating that there are no precisely-defined external goals and rewards for the agent. In a naturalistic context, we then have a situation where agents create actions based on the past history of actions and perceptions, and if there is any relevant notion of reward or goal, it is within the cognitive mechanism of some agent. A naturalistic agent X is then an agent 7 which is restricted to one particular naturalistic context, involving one particular environment pe (formally, we may achieve this within the framework of agents described above via dictating that X issues constant “null actions” ao in all environments except ;:). Next, we posit a metric space (2, d) of naturalistic agents defined on a naturalistic context involving environment yp, and a subspace A € &’, of inference agents, which are naturalistic agents that output predictions of other agents’ behaviors (a notion we will not fully formalize here). If agents are represented as program trees, then d may be taken as edit distance on tree space [Bil05]. Then, for each agent 6 € A, we may assess

Related Documents (6)

Dept. of JusticeOtherUnknown

EFTA Document EFTA01284096

KY(' Print Page I of 16 OB PWM GLOBAL KYC/NCA: PART A KYC Case r One sheet must be established per relationship - list all accounts included in the relationshi I. Relationship Details Relationship Nallle' SOUT14ERN FINANCIAL RELATIONSHIP:00030183290 a-mating Center Mar 'kirk Relationship Manager: Stewart Oldfiekl • New PWM Relationship 12 Existing PWM Relationship Relationship to PALM: II tasting, please indicate since when the relationship enists, provide mason for new profit a

16p
House OversightMar 1, 2013

Private Placement Memorandum for Knowledge Universe Education L.P. – $1 B Offering with Complex Offshore Structure, Related‑Party Deals, and Michael Milken Restrictions

Private Placement Memorandum for Knowledge Universe Education L.P. – $1 B Offering with Complex Offshore Structure, Related‑Party Deals, and Michael Milken Restrictions The document outlines a $1 billion private placement for a Cayman‑registered education partnership controlled by Michael and Lowell Milken and former ambassador Steven Green. It details extensive related‑party financing, high leverage, and a series of covenants that limit the principals’ ability to engage in certain businesses (e.g., Milken’s 1998 SEC injunction). While no new wrongdoing is disclosed, the structure raises red‑flag issues – offshore entities, mandatory conversion of high‑vote shares, and arbitration in London – that merit follow‑up for potential conflicts of interest, financial‑risk exposure, and regulatory compliance. Key insights: Offering of $1 billion (up to $1.5 billion) of Units backed by a Cayman exempted limited partnership (KUE).; Principal owners: Michael Milken, Lowell Milken, Steven Green – all with prior SEC and legal scrutiny (Milken’s 1998 judgment).; Complex capital structure: Common LP Units, Class A shares, Class B shares, Profits Participation Units, and high‑vote/low‑vote equity split.

1p
Dept. of JusticeOtherUnknown

EFTA Document EFTA01366403

prepared without regard to the disclosure standards for issuance prospectuses under art. 652a or art. 1156 of the Swiss Code of Obligations or the disclosure standards for listing prospectuses under art. 27 ff. of the SIX Listing Rules or the listing rules of any other stock exchange or regulated trading facility in Switzerland. Neither this document nor any other offering or marketing material relating to the shares or the offering may be publicly distributed or otherwise made publicly avai

1p
OtherUnknown

KYC Print

DOJ EFTA Data Set 10 document EFTA01284048

12p
Financial RecordUnknown

KYC Print

DOJ EFTA Data Set 10 document EFTA01283971

14p
Dept. of JusticeOtherUnknown

EFTA Document EFTA01284048

KYC Print Page I of 12 DB PWM GLOBAL KYC/NCA: PART A KYC Case a : 01346733 One sheet must be established per relationship - list all accounts Included in the relationship 1. Relationship Details Relationship Name: SW-DERN FINANCIAL RELATIONSHIP. Booking Center: New Tort Relationship Manager: Paul Monis O New OWN Relabonship O Existing PWM Relabonship Relabonship to PWM: If existing, please indicate since when the relationship exists, provide reason for new profile and attach old p

12p

Forum Discussions

This document was digitized, indexed, and cross-referenced with 1,500+ persons in the Epstein files. 100% free, ad-free, and independent.

Support This ProjectSupported by 1,550+ people worldwide
Annotations powered by Hypothesis. Select any text on this page to annotate or highlight it.