MétaCan
Menu
Back to cohort

Time-Division Online Speculative Inference for Cost-Effective Intelligent Internet of Things

2024· article· en· W4405909198 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.
fundA Canadian funder is recorded on the work.

Bibliographic record

Venuenot available
Typearticle
Languageen
FieldComputer Science
TopicAdvanced Decision-Making Techniques
Canadian institutionsUniversity of British Columbia
FundersPublic Safety Canada
KeywordsComputer scienceDivision (mathematics)InferenceInternet of ThingsThe InternetComputer networkComputer securityWorld Wide WebArtificial intelligence

Abstract

fetched live from OpenAlex

Intelligent Internet of Things (IIoT), a network paradigm, is an interconnection of intelligent edge devices, empowered by machine learning models. The recent emergence of large language models (LLMs) opens a new path towards IIoT. Although device models are generally lightweight and suitable for edge devices with limited processing power, their performance is less impressive compared to LLMs running on large servers. Server models offer superior performance with the cost of significant computational resources for model inference. To realize cost-effective IIoT, we introduce a novel collaborative inference framework: the online speculative inference enabled by device collaboration and time divisions. We aim to find the minimum number of servers required for the immediate processing of arrived inference requests. To solve the problem, we propose an evolving directed acyclic graph along with a Proof-of-Deletion mechanism for cost reduction and privacy protection. Based on computer simulations using heterogeneous LLMs, it is found that time-division online speculative inference is a promising approach towards cost-effective IIoT.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.001
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Other design · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.857
Threshold uncertainty score0.515

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.001
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.001
Open science0.0010.001
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.031
GPT teacher head0.374
Teacher spread0.343 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Quick stats

Citations0
Published2024
Admission routes2
Has abstractyes

Explore more

Same topicAdvanced Decision-Making TechniquesFrench-language works237,207