Emergence of Hidden Capabilities: Exploring Learning Dynamics in Concept Space

• URL: http://arxiv.org/abs/2406.19370v4
• Date: Wed, 11 Dec 2024 07:53:57 GMT
• Title: Emergence of Hidden Capabilities: Exploring Learning Dynamics in Concept Space
• Authors: Core Francisco Park, Maya Okawa, Andrew Lee, Hidenori Tanaka, Ekdeep Singh Lubana,
• Abstract summary: We analyze a model's learning dynamics via a framework we call the concept space.<n>We observe moments of sudden turns in the direction of a model's learning dynamics in concept space.<n>Surprisingly, these points precisely correspond to the emergence of hidden capabilities.
• Score: 14.188708813577456
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Modern generative models demonstrate impressive capabilities, likely stemming from an ability to identify and manipulate abstract concepts underlying their training data. However, fundamental questions remain: what determines the concepts a model learns, the order in which it learns them, and its ability to manipulate those concepts? To address these questions, we propose analyzing a model's learning dynamics via a framework we call the concept space, where each axis represents an independent concept underlying the data generating process. By characterizing learning dynamics in this space, we identify how the speed at which a concept is learned, and hence the order of concept learning, is controlled by properties of the data we term concept signal. Further, we observe moments of sudden turns in the direction of a model's learning dynamics in concept space. Surprisingly, these points precisely correspond to the emergence of hidden capabilities, i.e., where latent interventions show the model possesses the capability to manipulate a concept, but these capabilities cannot yet be elicited via naive input prompting. While our results focus on synthetically defined toy datasets, we hypothesize a general claim on emergence of hidden capabilities may hold: generative models possess latent capabilities that emerge suddenly and consistently during training, though a model might not exhibit these capabilities under naive input prompting.

Related papers

• FaCT: Faithful Concept Traces for Explaining Neural Network Decisions [56.796533084868884]
  Deep networks have shown remarkable performance across a wide range of tasks, yet getting a global concept-level understanding of how they function remains a key challenge.<n>We put emphasis on the faithfulness of concept-based explanations and propose a new model with model-inherent mechanistic concept-explanations.<n>Our concepts are shared across classes and, from any layer, their contribution to the logit and their input-visualization can be faithfully traced.
  arXiv  Detail & Related papers  (2025-10-29T13:35:46Z)
• Can Data-Driven Dynamics Reveal Hidden Physics? There Is A Need for Interpretable Neural Operators [10.591168773809635]
  We classify neural operators into two types: (1) Spatial domain models that learn on grids and (2) Functional domain models that learn with function bases.<n> Specifically, we provide a way to explain the prediction-making process of neural operators and show that neural operator can learn hidden physical patterns from data.<n>Next, we show that a simple dual-space multi-scale model can achieve SOTA performance and we believe that dual-space multi-spatio-scale models hold significant potential to learn complex physics.
  arXiv  Detail & Related papers  (2025-10-03T02:50:21Z)
• Concept-Guided Interpretability via Neural Chunking [64.6429903327095]
  We show that neural networks exhibit patterns in their raw population activity that mirror regularities in the training data.<n>We propose three methods to extract recurring chunks on a neural population level.<n>Our work points to a new direction for interpretability, one that harnesses both cognitive principles and the structure of naturalistic data.
  arXiv  Detail & Related papers  (2025-05-16T13:49:43Z)
• Meta-Representational Predictive Coding: Biomimetic Self-Supervised Learning [51.22185316175418]
  We present a new form of predictive coding that we call meta-representational predictive coding (MPC) MPC sidesteps the need for learning a generative model of sensory input by learning to predict representations of sensory input across parallel streams.
  arXiv  Detail & Related papers  (2025-03-22T22:13:14Z)
• I Predict Therefore I Am: Is Next Token Prediction Enough to Learn Human-Interpretable Concepts from Data? [79.01538178959726]
  Large language models (LLMs) have led many to conclude that they exhibit a form of intelligence. We introduce a novel generative model that generates tokens on the basis of human interpretable concepts represented as latent discrete variables.
  arXiv  Detail & Related papers  (2025-03-12T01:21:17Z)
• SOLD: Reinforcement Learning with Slot Object-Centric Latent Dynamics [16.020835290802548]
  Slot-Attention for Object-centric Latent Dynamics is a novel algorithm that learns object-centric dynamics models from pixel inputs. We demonstrate that the structured latent space not only improves model interpretability but also provides a valuable input space for behavior models to reason over. Our results show that SOLD outperforms DreamerV3, a state-of-the-art model-based RL algorithm, across a range of benchmark robotic environments.
  arXiv  Detail & Related papers  (2024-10-11T14:03:31Z)
• Learning Discrete Concepts in Latent Hierarchical Models [73.01229236386148]
  Learning concepts from natural high-dimensional data holds potential in building human-aligned and interpretable machine learning models. We formalize concepts as discrete latent causal variables that are related via a hierarchical causal model. We substantiate our theoretical claims with synthetic data experiments.
  arXiv  Detail & Related papers  (2024-06-01T18:01:03Z)
• Advancing Ante-Hoc Explainable Models through Generative Adversarial Networks [24.45212348373868]
  This paper presents a novel concept learning framework for enhancing model interpretability and performance in visual classification tasks. Our approach appends an unsupervised explanation generator to the primary classifier network and makes use of adversarial training. This work presents a significant step towards building inherently interpretable deep vision models with task-aligned concept representations.
  arXiv  Detail & Related papers  (2024-01-09T16:16:16Z)
• Interpreting Pretrained Language Models via Concept Bottlenecks [55.47515772358389]
  Pretrained language models (PLMs) have made significant strides in various natural language processing tasks. The lack of interpretability due to their black-box'' nature poses challenges for responsible implementation. We propose a novel approach to interpreting PLMs by employing high-level, meaningful concepts that are easily understandable for humans.
  arXiv  Detail & Related papers  (2023-11-08T20:41:18Z)
• Unraveling Feature Extraction Mechanisms in Neural Networks [10.13842157577026]
  We propose a theoretical approach based on Neural Tangent Kernels (NTKs) to investigate such mechanisms. We reveal how these models leverage statistical features during gradient descent and how they are integrated into final decisions. We find that while self-attention and CNN models may exhibit limitations in learning n-grams, multiplication-based models seem to excel in this area.
  arXiv  Detail & Related papers  (2023-10-25T04:22:40Z)
• Attributing Learned Concepts in Neural Networks to Training Data [5.930268338525991]
  We find evidence for convergence, where removing the 10,000 top attributing images for a concept and retraining the model does not change the location of the concept in the network. This suggests that the features that inform the development of a concept are spread in a more diffuse manner across its exemplars, implying robustness in concept formation.
  arXiv  Detail & Related papers  (2023-10-04T20:26:59Z)
• Exploring Model Transferability through the Lens of Potential Energy [78.60851825944212]
  Transfer learning has become crucial in computer vision tasks due to the vast availability of pre-trained deep learning models. Existing methods for measuring the transferability of pre-trained models rely on statistical correlations between encoded static features and task labels. We present an insightful physics-inspired approach named PED to address these challenges.
  arXiv  Detail & Related papers  (2023-08-29T07:15:57Z)
• Concept backpropagation: An Explainable AI approach for visualising learned concepts in neural network models [0.0]
  We present an extension to the method of concept detection, named emphconcept backpropagation, which provides a way of analysing how the information representing a given concept is internalised in a given neural network model.
  arXiv  Detail & Related papers  (2023-07-24T08:21:13Z)
• A Recursive Bateson-Inspired Model for the Generation of Semantic Formal Concepts from Spatial Sensory Data [77.34726150561087]
  This paper presents a new symbolic-only method for the generation of hierarchical concept structures from complex sensory data. The approach is based on Bateson's notion of difference as the key to the genesis of an idea or a concept. The model is able to produce fairly rich yet human-readable conceptual representations without training.
  arXiv  Detail & Related papers  (2023-07-16T15:59:13Z)
• On Binding Objects to Symbols: Learning Physical Concepts to Understand Real from Fake [155.6741526791004]
  We revisit the classic signal-to-symbol barrier in light of the remarkable ability of deep neural networks to generate synthetic data. We characterize physical objects as abstract concepts and use the previous analysis to show that physical objects can be encoded by finite architectures. We conclude that binding physical entities to digital identities is possible in finite time with finite resources.
  arXiv  Detail & Related papers  (2022-07-25T17:21:59Z)
• Discovering Concepts in Learned Representations using Statistical Inference and Interactive Visualization [0.76146285961466]
  Concept discovery is important for bridging the gap between non-deep learning experts and model end-users. Current approaches include hand-crafting concept datasets and then converting them to latent space directions. In this study, we offer another two approaches to guide user discovery of meaningful concepts, one based on multiple hypothesis testing, and another on interactive visualization.
  arXiv  Detail & Related papers  (2022-02-09T22:29:48Z)
• Deep Graph Memory Networks for Forgetting-Robust Knowledge Tracing [5.648636668261282]
  We propose a novel knowledge tracing model, namely emphDeep Graph Memory Network (DGMN) In this model, we incorporate a forget gating mechanism into an attention memory structure in order to capture forgetting behaviours. This model has the capability of learning relationships between latent concepts from a dynamic latent concept graph.
  arXiv  Detail & Related papers  (2021-08-18T12:04:10Z)
• Learning Stable Deep Dynamics Models [91.90131512825504]
  We propose an approach for learning dynamical systems that are guaranteed to be stable over the entire state space. We show that such learning systems are able to model simple dynamical systems and can be combined with additional deep generative models to learn complex dynamics.
  arXiv  Detail & Related papers  (2020-01-17T00:04:45Z)

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.