Semanlink - [1703.07464] No Fuss Distance Metric Learning using Proxies

[1703.07464] No Fuss Distance Metric Learning using Proxies

Tags:

> We address the problem of distance metric learning (DML), defined as learning a distance consistent with a notion of semantic similarity...
> Traditionnaly, supervision is expressed in the form of sets of points that follow
an ordinal relationship – an anchor point x is similar to
a set of positive points Y , and dissimilar to a set of negative
points Z, and a loss defined over these distances is minimized.
> Triplet-Based methods are challenging to optimize (a main issue is the need for finding informative triplets).
>
> We propose to **optimize the triplet loss on a different space of triplets, consisting of an anchor data point and similar and dissimilar proxy points which are learned as well**. These proxies approximate the original data points, so that a triplet loss over the proxies is a tight upper bound of the original loss.

Mentioned in this [blog post](/doc/2020/01/training_a_speaker_embedding_fr):

> "**Proxy based triplet learning**": instead of generating triplets, we learn an embedding for each class and use the learnt embedding as a proxy for triplets as part of the training. In other words, we can train end to end without the computationally expensive step of resampling triplets after each network update.

Near the conclusion:

> Our formulation of Proxy-NCA loss produces a loss very
similar to the standard cross-entropy loss used in classification.
However, we arrive at our formulation from a different
direction: we are not interested in the actual classifier and
indeed discard the proxies once the model has been trained.
Instead, the proxies are auxiliary variables, enabling more
effective optimization of the embedding model parameters.
**As such, our formulation not only enables us to surpass the
state of the art in zero-shot learning, but also offers an explanation
to the effectiveness of the standard trick of training
a classifier, and using its penultimate layer’s output as the
embedding.**

Au sujet de ce document

sl:arxiv_author :
sl:arxiv_firstAuthor : Yair Movshovitz-Attias
sl:arxiv_num : 1703.07464
sl:arxiv_published : 2017-03-21T23:11:56Z
sl:arxiv_summary : We address the problem of distance metric learning (DML), defined as learning a distance consistent with a notion of semantic similarity. Traditionally, for this problem supervision is expressed in the form of sets of points that follow an ordinal relationship -- an anchor point $x$ is similar to a set of positive points $Y$, and dissimilar to a set of negative points $Z$, and a loss defined over these distances is minimized. While the specifics of the optimization differ, in this work we collectively call this type of supervision Triplets and all methods that follow this pattern Triplet-Based methods. These methods are challenging to optimize. A main issue is the need for finding informative triplets, which is usually achieved by a variety of tricks such as increasing the batch size, hard or semi-hard triplet mining, etc. Even with these tricks, the convergence rate of such methods is slow. In this paper we propose to optimize the triplet loss on a different space of triplets, consisting of an anchor data point and similar and dissimilar proxy points which are learned as well. These proxies approximate the original data points, so that a triplet loss over the proxies is a tight upper bound of the original loss. This proxy-based loss is empirically better behaved. As a result, the proxy-loss improves on state-of-art results for three standard zero-shot learning datasets, by up to 15% points, while converging three times as fast as other triplet-based losses.@en
sl:arxiv_title : No Fuss Distance Metric Learning using Proxies@en
sl:arxiv_updated : 2017-08-01T19:52:13Z
sl:bookmarkOf : https://arxiv.org/abs/1703.07464
sl:creationDate : 2020-02-09
sl:creationTime : 2020-02-09T18:44:26Z
sl:relatedDoc : http://www.semanlink.net/doc/2020/01/training_a_speaker_embedding_fr

Infos sur le fichier

Bookmark of: https://arxiv.org/abs/1703.07464

Linked From

Training a Speaker Embedding from Scratch with Triplet Learning (2018)

Tags:

2020-01-23 A propos

Documents with similar tags (experimental)

[2304.02711] Structured prompt interrogation and recursive extraction of semantics (SPIRES): A method for populating knowledge bases using zero-shot learning

Tags:

2023-04-07 A propos

[2112.05682] Self-attention Does Not Need O(n^2) Memory

Tags:

2023-02-27 A propos

[2006.10713] Zero-Shot Learning with Common Sense Knowledge Graphs

Tags:

2022-08-29 A propos

[1712.05972] Train Once, Test Anywhere: Zero-Shot Learning for Text Classification

Tags:

2021-10-16 A propos

[2007.00849] Facts as Experts: Adaptable and Interpretable Neural Memory over Symbolic Knowledge

Tags:

> a neural language model that includes **an explicit interface between symbolically interpretable factual information and subsymbolic neural knowledge.**... **The model can be updated without re-training by manipulating its symbolic representations**. In particular this model allows us to add new facts and overwrite existing ones.

> a **neural language model which learns to access information
in a symbolic knowledge graph.**

> This
model builds on the recently-proposed [Entities as
Experts](doc:2020/07/2004_07202_entities_as_expert) (EaE) language model (Févry et al., 2020),
which extends the same transformer (Vaswani
et al., 2017) architecture of BERT (Devlin et al., 2019) with an additional external memory for entities.
>
> After training EaE, the embedding associated
with an entity will (ideally) capture information
about the textual context in which that
entity appears, and by inference, the entity’s semantic
properties
>
> we include an additional
memory called a fact memory, which encodes
triples from a symbolic KB.
>
> This combination results in a
neural language model which learns to access information
in a the symbolic knowledge graph.

TODO:

- read again IBM's [Span Selection Pre-training for Question Answering](doc:2019/09/_1909_04120_span_selection_pre) ("an effort to avoid encoding general knowledge in the transformer network itself")
- compare with [[1907.05242] Large Memory Layers with Product Keys](doc:2019/07/_1907_05242_large_memory_layer)
- how does it relate with [[2002.08909] REALM: Retrieval-Augmented Language Model Pre-Training](doc:2020/12/2002_08909_realm_retrieval_a)?

2020-07-09 A propos

[2006.15020] Pre-training via Paraphrasing

Tags: