Differentiable neural computer
artificial neural network architecture
In artificial intelligence, a differentiable neural computer (DNC) is a memory augmented neural network architecture (MANN), which is typically (but not by definition) recurrent in its implementation. The model was published in 2016 by Alex Graves et al. of DeepMind.
Applications
DNC indirectly takes inspiration from Von-Neumann architecture, making it likely to outperform conventional architectures in tasks that are fundamentally algorithmic that cannot be learned by finding a decision boundary.
So far, DNCs have been demonstrated to handle only relatively simple tasks, which can be solved using conventional programming. But DNCs don't need to be programmed for each problem, but can instead be trained. This attention span allows the user to feed complex data structures such as graphs sequentially, and recall them for later use. Furthermore, they can learn aspects of symbolic reasoning and apply it to working memory. The researchers who published the method see promise that DNCs can be trained to perform complex, structured tasks and address big-data applications that require some sort of reasoning, such as generating video commentaries or semantic text analysis.
DNC can be trained to navigate rapid transit systems, and apply that network to a different system.
The public source identifies “Differentiable neural computer” as artificial neural network architecture. This brief keeps that definition visible, then builds a research path around Differentiable, neural and computer.
Why this record matters
A short description can identify a subject without explaining its stakes. For “Differentiable neural computer”, the useful work is to connect “artificial neural network architecture” to the records capable of establishing context and consequence.
The strongest evidence will usually combine a dated visual record with documents produced by the authority responsible for the place. The source revision retrieved here is dated Aug 2, 2025. The linked authority identifier is Q28324912. None of the 0 selected statements returned an explicit reference. The first chronological checks are 2016.
A single description cannot reconcile every naming convention, rebuilding phase or administrative transfer associated with a place. The lead is largely declarative, so disagreement and counter-evidence require a deliberate search beyond the opening account. Authority statements aid reconciliation but still require their own references, qualifiers and ranks to be checked.
How to read it
Treat names, boundaries and functions as historically changeable. Maps, plans, inventories and administrative records can clarify what the place meant at different dates.
- Historic place names
- Jurisdictional context
- Routes into maps and plans
Contemporary maps, plans, listed-building records, estate papers and the responsible local or national archive.
Three-step research path
- Establish the record: confirm the title “Differentiable neural computer”, its source revision and the description used here.
- Expand the search: follow Differentiable neural computer primary sources, Differentiable neural computer archive and Differentiable research across catalogues and specialist indexes.
- Test the account: compare the strongest cited source with the responsible institution’s current record and note any disagreement.
Questions for further research
- Which source most directly establishes the central claim about “Differentiable neural computer”?
- Which earlier names or jurisdictions may reveal additional records?
- What physical evidence or contemporary plan supports the description?
Search terms from this dossier
This entry incorporates text from “Differentiable neural computer” on English Wikipedia. Contributors are listed in the page history. Text is available under the Creative Commons Attribution-ShareAlike 4.0 License. Selected authority identifiers and statements are retrieved from Wikidata under CC0; their references and qualifiers remain part of the verification path.