Pith. sign in

Object Representations as Fixed Points: Training Iterative Refinement Algorithms with Implicit Differentiation

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

Iterative refinement -- start with a random guess, then iteratively improve the guess -- is a useful paradigm for representation learning because it offers a way to break symmetries among equally plausible explanations for the data. This property enables the application of such methods to infer representations of sets of entities, such as objects in physical scenes, structurally resembling clustering algorithms in latent space. However, most prior works differentiate through the unrolled refinement process, which can make optimization challenging. We observe that such methods can be made differentiable by means of the implicit function theorem, and develop an implicit differentiation approach that improves the stability and tractability of training by decoupling the forward and backward passes. This connection enables us to apply advances in optimizing implicit layers to not only improve the optimization of the slot attention module in SLATE, a state-of-the-art method for learning entity representations, but do so with constant space and time complexity in backpropagation and only one additional line of code.

fields

cs.LG 1

years

2025 1

verdicts

REJECT 1

representative citing papers

Identifiable Object Representations under Spatial Ambiguities

cs.LG · 2025-06-09 · reject · novelty 6.0

VISA learns view-invariant object representations by aggregating probabilistic slots across multiple unlabeled viewpoints, with an identifiability analysis up to affine and permutation equivalence.

citing papers explorer

Showing 1 of 1 citing paper.

  • Identifiable Object Representations under Spatial Ambiguities cs.LG · 2025-06-09 · reject · none · ref 11 · internal anchor

    VISA learns view-invariant object representations by aggregating probabilistic slots across multiple unlabeled viewpoints, with an identifiability analysis up to affine and permutation equivalence.