Causal and temporal reasoning about video dynamics is a challenging problem. While neuro-symbolic models that combine symbolic reasoning with neural-based perception and prediction have shown promise, they exhibit limitations, especially in answering counterfactual questions. This paper introduces a method to enhance a neuro-symbolic model for counterfactual reasoning, leveraging symbolic reasoning about causal relations among events. We define the notion of a causal graph to represent such relations and use Answer Set Programming (ASP), a declarative logic programming method, to find how to coordinate perception and simulation modules. We validate the effectiveness of our approach on two benchmarks, CLEVRER and CRAFT. Our enhancement achieves state-of-the-art performance on the CLEVRER challenge, significantly outperforming existing models. In the case of the CRAFT benchmark, we leverage a large pre-trained language model, such as GPT-3.5 and GPT-4, as a proxy for a dynamics simulator. Our findings show that this method can further improve its performance on counterfactual questions by providing alternative prompts instructed by symbolic causal reasoning.

In recent years, researchers introduced the Multi-Agent Path Finding in Continuous Time (MAPFR) problem. Conflict-based search with Continuous Time (CCBS), a variant of CBS for discrete MAPF, aims to solve MAPFR with completeness and optimality guarantees. However, CCBS overlooked the fact that search algorithms only guarantee termination and return the optimal solution with a finite amount of search nodes. In this paper, we show that CCBS is incomplete, reveal the gaps in the existing implementation, demonstrate that patching is non-trivial, and discuss the next steps.

Motion planning against sensor data is often a critical bottleneck in real-time robot control. For sampling-based motion planners, which are effective for high-dimensional systems such as manipulators, the most time-intensive component is collision checking. We present a novel spatial data structure, the collision-affording point tree (CAPT): an exact representation of point clouds that accelerates collision-checking queries between robots and point clouds by an order of magnitude, with an average query time of less than 10 nanoseconds on 3D scenes comprising thousands of points. With the CAPT, sampling-based planners can generate valid, high-quality paths in under a millisecond, with total end-to-end computation time faster than 60 FPS, on a single thread of a consumer-grade CPU. We also present a point cloud filtering algorithm, based on space-filling curves, which reduces the number of points in a point cloud while preserving structure. Our approach enables robots to plan at real-time speeds in sensed environments, opening up potential uses of planning for high-dimensional systems in dynamic, changing, and unmodeled environments.

Weight-sharing is ubiquitous in deep learning. Motivated by this, we introduce ''weight-sharing regularization'' for neural networks, defined as $R(w) = \frac{1}{d - 1}\sum_{i > j}^d |w_i - w_j|$. We study the proximal mapping of $R$ and provide an intuitive interpretation of it in terms of a physical system of interacting particles. Using this interpretation, we design a novel parallel algorithm for $\operatorname{prox}_R$ which provides an exponential speedup over previous algorithms, with a depth of $O(\log^3 d)$. Our algorithm makes it feasible to train weight-sharing regularized deep neural networks with proximal gradient descent. Experiments reveal that weight-sharing regularization enables fully-connected networks to learn convolution-like filters.

One way of ensuring operator's safety during human-robot collaboration is through Speed and Separation Monitoring (SSM), as defined in ISO standard ISO/TS 15066. In general, it is impossible to avoid all human-robot collisions: consider for instance the case when the robot does not move at all, a human operator can still collide with it by hitting it of her own voluntary motion. In the SSM framework, it is possible however to minimize harm by requiring this: \emph{if} a collision ever occurs, then the robot must be in a \emph{stationary state} (all links have zero velocity) at the time instant of the collision. In this paper, we propose a time-optimal control policy based on Time-Optimal Path Parameterization (TOPP) to guarantee such a behavior. Specifically, we show that: for any robot motion that is strictly faster than the motion recommended by our policy, there exists a human motion that results in a collision with the robot in a non-stationary state. Correlatively, we show, in simulation, that our policy is strictly less conservative than state-of-the-art safe robot control methods. Additionally, we propose a parallelization method to reduce the computation time of our pre-computation phase (down to 0.5 sec, practically), which enables the whole pipeline (including the pre-computation) to be executed at runtime, nearly in real-time. Finally, we demonstrate the application of our method in a scenario: time-optimal, safe control of a 6-dof industrial robot.

A critical task for developing safe autonomous driving stacks is to determine whether an obstacle is safety-critical, i.e., poses an imminent threat to the autonomous vehicle. Our previous work showed that Hamilton Jacobi reachability theory can be applied to compute interaction-dynamics-aware perception safety zones that better inform an ego vehicle's perception module which obstacles are considered safety-critical. For completeness, these zones are typically larger than absolutely necessary, forcing the perception module to pay attention to a larger collection of objects for the sake of conservatism. As an improvement, we propose a maneuver-based decomposition of our safety zones that leverages information about the ego maneuver to reduce the zone volume. In particular, we propose a "temporal convolution" operation that produces safety zones for specific ego maneuvers, thus limiting the ego's behavior to reduce the size of the safety zones. We show with numerical experiments that maneuver-based zones are significantly smaller (up to 76% size reduction) than the baseline while maintaining completeness.

Multimodal text-to-image models, like DALL-Large language models trained for code generation can be E 2 [34], Midjourney [11] or Stable Diffusion [35] are applied to speaking virtual worlds into existence (creating raising concerns about displacing concept artists and have virtual worlds). In this work we show that prompt-based already won at least one major art competition [36]. Large methods can both accelerate in-VR level editing, as well Language Models (LLMs), like GPT-3 [6], are not only as can become part of gameplay rather than just part of generating very convincing text completions, but have game development. As an example, we present Codex recently become capable of generating code with models VR Pong which shows non-deterministic game mechanics like OpenAI Codex [8] or AlphaCode [25]. We propose using generative processes to not only create static content in this paper that these capabilities can be combined to but also non-trivial interactions between 3D objects. This allow "speaking the world into existence", or taking natural demonstration naturally leads to an integral discussion on language descriptions and turning them into interactive how one would evaluate and benchmark experiences created visual scenes within a game engine. In particular, this by generative models - as there are no qualitative or has the potential for allowing authoring Virtual Reality quantitative metrics that apply in these scenarios. We conclude (VR) experiences from within the headset, as well as allow by discussing impending challenges of AI-assisted completely novel modes of gameplay.

Dialog systems must be capable of incorporating new skills via updates over time in order to reflect new use cases or deployment scenarios. Similarly, developers of such ML-driven systems need to be able to add new training data to an already-existing dataset to support these new skills. In intent classification systems, problems can arise if training data for a new skill's intent overlaps semantically with an already-existing intent. We call such cases collisions. This paper introduces the task of intent collision detection between multiple datasets for the purposes of growing a system's skillset. We introduce several methods for detecting collisions, and evaluate our methods on real datasets that exhibit collisions. To highlight the need for intent collision detection, we show that model performance suffers if new data is added in such a way that does not arbitrate colliding intents. Finally, we use collision detection to construct and benchmark a new dataset, Redwood, which is composed of 451 ntent categories from 13 original intent classification datasets, making it the largest publicly available intent classification benchmark.