Optimization Encoders: Rethinking Second-Order Meta-Learning for Neural Fields

Key points

• Conditional neural fields represent signals continuously, but their effectiveness depends on how the conditional latent representations are inferred from observed data.1
• In meta-learning, this encoding occurs through gradient updates induced by the decoder, tying representation learning directly to decoder design.1
• We formalize this connection by interpreting latent optimization as an optimization encoder, unifying the roles of second-order differentiation, latent parameterization, and task supervision.1
• Disentangling the inner encoding objective from outer task supervision unifies reconstruction, classification, and segmentation within an end-to-end meta-learning framework, using reconstruction-only latent adaptation at test time.1

Sources (1)

• [1]Optimization Encoders: Rethinking Second-Order Meta-Learning for Neural Fields
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 6, 10:05 AM

  Guided by this view, we introduce Attentive Latent Fields (MetaLF), an equivariant transformer-based neural field that contextualizes a latent pointcloud through self-attention.

  Conditional neural fields represent signals continuously, but their effectiveness depends on how the conditional latent representations are inferred from observed data.

Extractive summary: sentences quoted from the sources.