Rethinking Vision Transformer for Object Centric Foundation Models

Recent state-of-the-art object segmentation mechanisms, such as the Segment Anything Model (SAM) and FastSAM, first encode the full image over several layers and then focus on generating the mask for one particular object or area. We present an off-grid Fovea-Like Input Patching (FLIP) approach, which selects image input and encodes it from the beginning in an object-focused manner. While doing so, it separates locational encoding from an object-centric perceptual code. FLIP is more data-efficient and yields improved segmentation performance when masking relatively small objects in high-resolution visual scenes. On standard benchmarks such as Hypersim, KITTI-360, and OpenImages, FLIP achieves Intersection over Union (IoU) scores that approach the performance of SAM with much less compute effort. It surpasses FastSAM in all IoU measurements. We also introduce an additional semi-natural but highly intuitive dataset where FLIP outperforms SAM and FastSAM overall and particularly on relatively small objects. Seeing that FLIP is an end-to-end object-centric segmentation approach, it has high potential particularly for applications that benefit from computationally efficient, spatially highly selective object tracking.

Quick and Accurate Affordance Learning

Infants learn actively in their environments, shaping their own learning curricula. They learn about their environments' affordances, that is, how local circumstances determine how their behavior can affect the environment. Here we model this type of behavior by means of a deep learning architecture. The architecture mediates between global cognitive map exploration and local affordance learning. Inference processes actively move the simulated agent towards regions where they expect affordance-related knowledge gain. We contrast three measures of uncertainty to guide this exploration: predicted uncertainty of a model, standard deviation between the means of several models (SD), and the Jensen-Shannon Divergence (JSD) between several models. We show that the first measure gets fooled by aleatoric uncertainty inherent in the environment, while the two other measures focus learning on epistemic uncertainty. JSD exhibits the most balanced exploration strategy. From a computational perspective, our model suggests three key ingredients for coordinating the active generation of learning curricula: (1) Navigation behavior needs to be coordinated with local motor behavior for enabling active affordance learning. (2) Affordances need to be encoded locally for acquiring generalized knowledge. (3) Effective active affordance learning mechanisms should use density comparison techniques for estimating expected knowledge gain. Future work may seek collaborations with developmental psychology to model active play in children in more realistic scenarios.

Loci-Segmented: Improving Scene Segmentation Learning

Slot-oriented processing approaches for compositional scene representation have recently undergone a tremendous development. We present Loci-Segmented (Loci-s), an advanced scene segmentation neural network that extends the slot-based location and identity tracking architecture Loci (Traub et al., ICLR 2023). The main advancements are (i) the addition of a pre-trained dynamic background module; (ii) a hyper-convolution encoder module, which enables object-focused bottom-up processing; and (iii) a cascaded decoder module, which successively generates object masks, masked depth maps, and masked, depth-map-informed RGB reconstructions. The background module features the learning of both a foreground identifying module and a background re-generator. We further improve performance via (a) the integration of depth information as well as improved slot assignments via (b) slot-location-entity regularization and (b) a prior segmentation network. Even without these latter improvements, the results reveal superior segmentation performance in the MOVi datasets and in another established dataset collection. With all improvements, Loci-s achieves a 32% better intersection over union (IoU) score in MOVi-E than the previous best. We furthermore show that Loci-s generates well-interpretable latent representations. We believe that these representations may serve as a foundation-model-like interpretable basis for solving downstream tasks, such as grounding language and context- and goal-conditioned event processing.

Looping LOCI: Developing Object Permanence from Videos

Recent compositional scene representation learning models have become remarkably good in segmenting and tracking distinct objects within visual scenes. Yet, many of these models require that objects are continuously, at least partially, visible. Moreover, they tend to fail on intuitive physics tests, which infants learn to solve over the first months of their life. Our goal is to advance compositional scene representation algorithms with an embedded algorithm that fosters the progressive learning of intuitive physics, akin to infant development. As a fundamental component for such an algorithm, we introduce Loci-Looped, which advances a recently published unsupervised object location, identification, and tracking neural network architecture (Loci, Traub et al., ICLR 2023) with an internal processing loop. The loop is designed to adaptively blend pixel-space information with anticipations yielding information-fused activities as percepts. Moreover, it is designed to learn compositional representations of both individual object dynamics and between-objects interaction dynamics. We show that Loci-Looped learns to track objects through extended periods of object occlusions, indeed simulating their hidden trajectories and anticipating their reappearance, without the need for an explicit history buffer. We even find that Loci-Looped surpasses state-of-the-art models on the ADEPT and the CLEVRER dataset, when confronted with object occlusions or temporary sensory data interruptions. This indicates that Loci-Looped is able to learn the physical concepts of object permanence and inertia in a fully unsupervised emergent manner. We believe that even further architectural advancements of the internal loop - also in other compositional scene representation learning models - can be developed in the near future.

Inductive biases in deep learning models for weather prediction

Deep learning has recently gained immense popularity in the Earth sciences as it enables us to formulate purely data-driven models of complex Earth system processes. Deep learning-based weather prediction (DLWP) models have made significant progress in the last few years, achieving forecast skills comparable to established numerical weather prediction (NWP) models with comparatively lesser computational costs. In order to train accurate, reliable, and tractable DLWP models with several millions of parameters, the model design needs to incorporate suitable inductive biases that encode structural assumptions about the data and modelled processes. When chosen appropriately, these biases enable faster learning and better generalisation to unseen data. Although inductive biases play a crucial role in successful DLWP models, they are often not stated explicitly and how they contribute to model performance remains unclear. Here, we review and analyse the inductive biases of six state-of-the-art DLWP models, involving a deeper look at five key design elements: input data, forecasting objective, loss components, layered design of the deep learning architectures, and optimisation methods. We show how the design choices made in each of the five design elements relate to structural assumptions. Given recent developments in the broader DL community, we anticipate that the future of DLWP will likely see a wider use of foundation models -- large models pre-trained on big databases with self-supervised learning -- combined with explicit physics-informed inductive biases that allow the models to provide competitive forecasts even at the more challenging subseasonal-to-seasonal scales.

Intelligent problem-solving as integrated hierarchical reinforcement learning

According to cognitive psychology and related disciplines, the development of complex problem-solving behaviour in biological agents depends on hierarchical cognitive mechanisms. Hierarchical reinforcement learning is a promising computational approach that may eventually yield comparable problem-solving behaviour in artificial agents and robots. However, to date the problem-solving abilities of many human and non-human animals are clearly superior to those of artificial systems. Here, we propose steps to integrate biologically inspired hierarchical mechanisms to enable advanced problem-solving skills in artificial agents. Therefore, we first review the literature in cognitive psychology to highlight the importance of compositional abstraction and predictive processing. Then we relate the gained insights with contemporary hierarchical reinforcement learning methods. Interestingly, our results suggest that all identified cognitive mechanisms have been implemented individually in isolated computational architectures, raising the question of why there exists no single unifying architecture that integrates them. As our final contribution, we address this question by providing an integrative perspective on the computational challenges to develop such a unifying architecture. We expect our results to guide the development of more sophisticated cognitively inspired hierarchical machine learning architectures.

Developing hierarchical anticipations via neural network-based event segmentation

Humans can make predictions on various time scales and hierarchical levels. Thereby, the learning of event encodings seems to play a crucial role. In this work we model the development of hierarchical predictions via autonomously learned latent event codes. We present a hierarchical recurrent neural network architecture, whose inductive learning biases foster the development of sparsely changing latent state that compress sensorimotor sequences. A higher level network learns to predict the situations in which the latent states tend to change. Using a simulated robotic manipulator, we demonstrate that the system (i) learns latent states that accurately reflect the event structure of the data, (ii) develops meaningful temporal abstract predictions on the higher level, and (iii) generates goal-anticipatory behavior similar to gaze behavior found in eye-tracking studies with infants. The architecture offers a step towards autonomous, self-motivated learning of compressed hierarchical encodings of gathered experiences and the exploitation of these encodings for the generation of highly versatile, adaptive behavior.

Binding Dancers Into Attractors

To effectively perceive and process observations in our environment, feature binding and perspective taking are crucial cognitive abilities. Feature binding combines observed features into one entity, called a Gestalt. Perspective taking transfers the percept into a canonical, observer-centered frame of reference. Here we propose a recurrent neural network model that solves both challenges. We first train an LSTM to predict 3D motion dynamics from a canonical perspective. We then present similar motion dynamics with novel viewpoints and feature arrangements. Retrospective inference enables the deduction of the canonical perspective. Combined with a robust mutual-exclusive softmax selection scheme, random feature arrangements are reordered and precisely bound into known Gestalt percepts. To corroborate evidence for the architecture's cognitive validity, we examine its behavior on the silhouette illusion, which elicits two competitive Gestalt interpretations of a rotating dancer. Our system flexibly binds the information of the rotating figure into the alternative attractors resolving the illusion's ambiguity and imagining the respective depth interpretation and the corresponding direction of rotation. We finally discuss the potential universality of the proposed mechanisms.

Learning What and Where -- Unsupervised Disentangling Location and Identity Tracking

Our brain can almost effortlessly decompose visual data streams into background and salient objects. Moreover, it can track the objects and anticipate their motion and interactions. In contrast, recent object reasoning datasets, such as CATER, have revealed fundamental shortcomings of current vision-based AI systems, particularly when targeting explicit object encodings, object permanence, and object reasoning. We introduce an unsupervised disentangled LOCation and Identity tracking system (Loci), which excels on the CATER tracking challenge. Inspired by the dorsal-ventral pathways in the brain, Loci tackles the what-and-where binding problem by means of a self-supervised segregation mechanism. Our autoregressive neural network partitions and distributes the visual input stream across separate, identically-parameterized and autonomously recruited neural network modules. Each module binds what with where, that is, compressed Gestalt encodings with locations. On the deep latent encoding levels interaction dynamics are processed. Besides exhibiting superior performance in current benchmarks, we propose that Loci may set the stage for deeper, explanation-oriented video processing -- akin to some deeper networked processes in the brain that appear to integrate individual entity and spatiotemporal interaction dynamics into event structures.

Inference of Affordances and Active Motor Control in Simulated Agents

Flexible, goal-directed behavior is a fundamental aspect of human life. Based on the free energy minimization principle, the theory of active inference formalizes the generation of such behavior from a computational neuroscience perspective. Based on the theory, we introduce an output-probabilistic, temporally predictive, modular artificial neural network architecture, which processes sensorimotor information, infers behavior-relevant aspects of its world, and invokes highly flexible, goal-directed behavior. We show that our architecture, which is trained end-to-end to minimize an approximation of free energy, develops latent states that can be interpreted as affordance maps. That is, the emerging latent states signal which actions lead to which effects dependent on the local context. In combination with active inference, we show that flexible, goal-directed behavior can be invoked, incorporating the emerging affordance maps. As a result, our simulated agent flexibly steers through continuous spaces, avoids collisions with obstacles, and prefers pathways that lead to the goal with high certainty. Additionally, we show that the learned agent is highly suitable for zero-shot generalization across environments: After training the agent in a handful of fixed environments with obstacles and other terrains affecting its behavior, it performs similarly well in procedurally generated environments containing different amounts of obstacles and terrains of various sizes at different locations. To improve and focus model learning further, we plan to invoke active inference-based, information-gain-oriented behavior also while learning the temporally predictive model itself in the near future. Moreover, we intend to foster the development of both deeper event-predictive abstractions and compact, habitual behavioral primitives.