Representational Multiplicity Should Be Exposed, Not Eliminated
- URL: http://arxiv.org/abs/2206.08890v1
- Date: Fri, 17 Jun 2022 16:53:12 GMT
- Title: Representational Multiplicity Should Be Exposed, Not Eliminated
- Authors: Ari Heljakka, Martin Trapp, Juho Kannala, Arno Solin
- Abstract summary: Two machine learning models with similar performance during training can have very different real-world performance characteristics.
This implies elusive differences in the internals of the models, manifesting as representational multiplicity (RM)
We introduce a conceptual and experimental setup for analyzing RM and show that certain training methods systematically result in greater RM than others.
- Score: 27.495944788838457
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: It is prevalent and well-observed, but poorly understood, that two machine
learning models with similar performance during training can have very
different real-world performance characteristics. This implies elusive
differences in the internals of the models, manifesting as representational
multiplicity (RM). We introduce a conceptual and experimental setup for
analyzing RM and show that certain training methods systematically result in
greater RM than others, measured by activation similarity via singular vector
canonical correlation analysis (SVCCA). We further correlate it with predictive
multiplicity measured by the variance in i.i.d. and out-of-distribution test
set predictions, in four common image data sets. We call for systematic
measurement and maximal exposure, not elimination, of RM in models. Qualitative
tools such as our confabulator analysis can facilitate understanding and
communication of RM effects to stakeholders.
Related papers
- Zero-Shot Embeddings Inform Learning and Forgetting with Vision-Language Encoders [6.7181844004432385]
The Inter-Intra Modal Measure (IIMM) functions as a strong predictor of performance changes with fine-tuning.
Fine-tuning on tasks with higher IIMM scores produces greater in-domain performance gains but also induces more severe out-of-domain performance degradation.
With only a single forward pass of the target data, practitioners can leverage this key insight to evaluate the degree to which a model can be expected to improve following fine-tuning.
arXiv Detail & Related papers (2024-07-22T15:35:09Z) - Revealing Multimodal Contrastive Representation Learning through Latent
Partial Causal Models [85.67870425656368]
We introduce a unified causal model specifically designed for multimodal data.
We show that multimodal contrastive representation learning excels at identifying latent coupled variables.
Experiments demonstrate the robustness of our findings, even when the assumptions are violated.
arXiv Detail & Related papers (2024-02-09T07:18:06Z) - Mitigating Biases with Diverse Ensembles and Diffusion Models [99.6100669122048]
We propose an ensemble diversification framework exploiting Diffusion Probabilistic Models (DPMs)
We show that DPMs can generate images with novel feature combinations, even when trained on samples displaying correlated input features.
We show that DPM-guided diversification is sufficient to remove dependence on primary shortcut cues, without a need for additional supervised signals.
arXiv Detail & Related papers (2023-11-23T15:47:33Z) - Leveraging Diffusion Disentangled Representations to Mitigate Shortcuts
in Underspecified Visual Tasks [92.32670915472099]
We propose an ensemble diversification framework exploiting the generation of synthetic counterfactuals using Diffusion Probabilistic Models (DPMs)
We show that diffusion-guided diversification can lead models to avert attention from shortcut cues, achieving ensemble diversity performance comparable to previous methods requiring additional data collection.
arXiv Detail & Related papers (2023-10-03T17:37:52Z) - Using Explainable Boosting Machine to Compare Idiographic and Nomothetic
Approaches for Ecological Momentary Assessment Data [2.0824228840987447]
This paper explores the use of non-linear interpretable machine learning (ML) models in classification problems.
Various ensembles of trees are compared to linear models using imbalanced synthetic and real-world datasets.
In one of the two real-world datasets, knowledge distillation method achieves improved AUC scores.
arXiv Detail & Related papers (2022-04-04T17:56:37Z) - An empirical evaluation of attention-based multi-head models for
improved turbofan engine remaining useful life prediction [9.282239595143787]
A single unit (head) is the conventional input feature extractor in deep learning architectures trained on multivariate time series signals.
This work extends the conventional single-head deep learning models to a more robust form by developing context-specific heads.
arXiv Detail & Related papers (2021-09-04T01:13:47Z) - Multi-Agent Imitation Learning with Copulas [102.27052968901894]
Multi-agent imitation learning aims to train multiple agents to perform tasks from demonstrations by learning a mapping between observations and actions.
In this paper, we propose to use copula, a powerful statistical tool for capturing dependence among random variables, to explicitly model the correlation and coordination in multi-agent systems.
Our proposed model is able to separately learn marginals that capture the local behavioral patterns of each individual agent, as well as a copula function that solely and fully captures the dependence structure among agents.
arXiv Detail & Related papers (2021-07-10T03:49:41Z) - Mean Embeddings with Test-Time Data Augmentation for Ensembling of
Representations [8.336315962271396]
We look at the ensembling of representations and propose mean embeddings with test-time augmentation (MeTTA)
MeTTA significantly boosts the quality of linear evaluation on ImageNet for both supervised and self-supervised models.
We believe that spreading the success of ensembles to inference higher-quality representations is the important step that will open many new applications of ensembling.
arXiv Detail & Related papers (2021-06-15T10:49:46Z) - Improving the Reconstruction of Disentangled Representation Learners via Multi-Stage Modeling [55.28436972267793]
Current autoencoder-based disentangled representation learning methods achieve disentanglement by penalizing the ( aggregate) posterior to encourage statistical independence of the latent factors.
We present a novel multi-stage modeling approach where the disentangled factors are first learned using a penalty-based disentangled representation learning method.
Then, the low-quality reconstruction is improved with another deep generative model that is trained to model the missing correlated latent variables.
arXiv Detail & Related papers (2020-10-25T18:51:15Z) - Machine learning for causal inference: on the use of cross-fit
estimators [77.34726150561087]
Doubly-robust cross-fit estimators have been proposed to yield better statistical properties.
We conducted a simulation study to assess the performance of several estimators for the average causal effect (ACE)
When used with machine learning, the doubly-robust cross-fit estimators substantially outperformed all of the other estimators in terms of bias, variance, and confidence interval coverage.
arXiv Detail & Related papers (2020-04-21T23:09:55Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.