In the rapidly evolving domain of vision-based deep reinforcement learning (RL), a pivotal challenge is to achieve generalization capability to dynamic environmental changes reflected in visual observations.Our work delves into the intricacies of this problem, identifying two key issues that appear in previous approaches for visual RL generalization: (i) imbalanced saliency and (ii) observational overfitting.Imbalanced saliency is a phenomenon where an RL agent disproportionately identifies salient features across consecutive frames in a frame stack. Observational overfitting occurs when the agent focuses on certain background regions rather than task-relevant objects.To address these challenges, we present a simple yet effective framework for generalization in visual RL (SimGRL) under dynamic scene perturbations.First, to mitigate the imbalanced saliency problem, we introduce an architectural modification to the image encoder to stack frames at the feature level rather than the image level.Simultaneously, to alleviate the observational overfitting problem, we propose a novel technique called shifted random overlay augmentation, which is specifically designed to learn robust representations capable of effectively handling dynamic visual scenes.Extensive experiments demonstrate the superior generalization capability of SimGRL, achieving state-of-the-art performance in benchmarks including the DeepMind Control Suite.

Visual grounding is a common vision task that involves grounding descriptive sentences to the corresponding regions of an image. Most existing methods use independent image-text encoding and apply complex hand-crafted modules or encoder-decoder architectures for modal interaction and query reasoning. However, their performance significantly drops when dealing with complex textual expressions. This is because the former paradigm only utilizes limited downstream data to fit the multi-modal feature fusion. Therefore, it is only effective when the textual expressions are relatively simple. In contrast, given the wide diversity of textual expressions and the uniqueness of downstream training data, the existing fusion module, which extracts multimodal content from a visual-linguistic context, has not been fully investigated.

Domain generalization (DG) methods aim to maintain good performance in an unseen target domain by using training data from multiple source domains. While success on certain occasions are observed, enhancing the baseline across most scenarios remains challenging. This work introduces a simple yet effective framework, dubbed learning from multiple experts (LFME), that aims to make the target model an expert in all source domains to improve DG. Specifically, besides learning the target model used in inference, LFME will also train multiple experts specialized in different domains, whose output probabilities provide professional guidance by simply regularizing the logit of the target model. Delving deep into the framework, we reveal that the introduced logit regularization term implicitly provides effects of enabling the target model to harness more information, and mining hard samples from the experts during training.

Dynamic jumping on high platforms and over gaps differentiates legged robots from wheeled counterparts. Compared to walking on rough terrains, dynamic locomotion on abrupt surfaces requires fusing proprioceptive and exteroceptive perception for explosive movements. In this paper, we propose SF-TIM (Simple Framework combining Terrain Imagination and Measurement), a single-policy method that enhances quadrupedal robot jumping agility, while preserving their fundamental blind walking capabilities. In addition, we introduce a terrain-guided reward design specifically to assist quadrupedal robots in high jumping, improving their performance in this task. To narrow the simulation-to-reality gap in quadrupedal robot learning, we introduce a stable and high-speed elevation map generation framework, enabling zero-shot simulation-to-reality transfer of locomotion ability. Our algorithm has been deployed and validated on both the small-/large-size quadrupedal robots, demonstrating its effectiveness in real-world applications: the robot has successfully traversed various high platforms and gaps, showing the robustness of our proposed approach. A demo video has been made available at https://flysoaryun.github.io/SF-TIM.

Deep neural networks have exhibited remarkable performance in various domains. However, the reliance of these models on spurious features has raised concerns about their reliability. A promising solution to this problem is last-layer retraining, which involves retraining the linear classifier head on a small subset of data without spurious cues. Nevertheless, selecting this subset requires human supervision, which reduces its scalability. Moreover, spurious cues may still exist in the selected subset. As a solution to this problem, we propose a novel ranking framework that leverages an open vocabulary object detection technique to identify images without spurious cues. More specifically, we use the object detector as a measure to score the presence of the target object in the images. Next, the images are sorted based on this score, and the last-layer of the model is retrained on a subset of the data with the highest scores. Our experiments on the ImageNet-1k dataset demonstrate the effectiveness of this ranking framework in sorting images based on spuriousness and using them for last-layer retraining.

In this paper, we present a simple framework of skill transfer learning for robotic ultrasound-guidance procedures. We briefly review challenges in skill transfer learning for robotic ultrasound-guidance procedures. We then identify the need of appropriate sampling techniques, computationally efficient neural networks models that lead to the proposal of a simple framework of skill transfer learning for real-time applications in robotic ultrasound-guidance procedures. We present pilot experiments from two participants (one experienced clinician and one non-clinician) looking for an optimal scanning plane of the four-chamber cardiac view from a fetal phantom. We analysed ultrasound image frames, time series of texture image features and quaternions and found that the experienced clinician performed the procedure in a quicker and smoother way compared to lengthy and non-constant movements from non-clinicians. For future work, we pointed out the need of pruned and quantised neural network models for real-time applications in robotic ultrasound-guidance procedure. The resources to reproduce this work are available at \url{https://github.com/mxochicale/rami-icra2023}.

I recently started an AI-focused educational newsletter, that already has over 150,000 subscribers. TheSequence is a no-BS (meaning no hype, no news etc) ML-oriented newsletter that takes 5 minutes to read. The goal is to keep you up to date with machine learning projects, research papers and concepts. Fine-tuning text-to-image(TTI) models are becoming an increasingly important aspect of applying these techniques in real-world scenarios. Despite the progress with TTI models, fine-tuning remains relatively challenging.

Graph contrastive learning (GCL) has emerged as a dominant technique for graph representation learning which maximizes the mutual information between paired graph augmentations that share the same semantics. Unfortunately, it is difficult to preserve semantics well during augmentations in view of the diverse nature of graph data. Currently, data augmentations in GCL that are designed to preserve semantics broadly fall into three unsatisfactory ways. First, the augmentations can be manually picked per dataset by trial-and-errors. Second, the augmentations can be selected via cumbersome search. Third, the augmentations can be obtained by introducing expensive domain-specific knowledge as guidance. All of these limit the efficiency and more general applicability of existing GCL methods. To circumvent these crucial issues, we propose a \underline{Sim}ple framework for \underline{GRA}ph \underline{C}ontrastive l\underline{E}arning, \textbf{SimGRACE} for brevity, which does not require data augmentations. Specifically, we take original graph as input and GNN model with its perturbed version as two encoders to obtain two correlated views for contrast. SimGRACE is inspired by the observation that graph data can preserve their semantics well during encoder perturbations while not requiring manual trial-and-errors, cumbersome search or expensive domain knowledge for augmentations selection. Also, we explain why SimGRACE can succeed. Furthermore, we devise adversarial training scheme, dubbed \textbf{AT-SimGRACE}, to enhance the robustness of graph contrastive learning and theoretically explain the reasons. Albeit simple, we show that SimGRACE can yield competitive or better performance compared with state-of-the-art methods in terms of generalizability, transferability and robustness, while enjoying unprecedented degree of flexibility and efficiency.

The research process includes many decisions, e.g., how to entitle and where to publish the paper. In this paper, we introduce a general framework for investigating the effects of such decisions. The main difficulty in investigating the effects is that we need to know counterfactual results, which are not available in reality. The key insight of our framework is inspired by the existing counterfactual analysis using twins, where the researchers regard twins as counterfactual units. The proposed framework regards a pair of papers that cite each other as twins. Such papers tend to be parallel works, on similar topics, and in similar communities. We investigate twin papers that adopted different decisions, observe the progress of the research impact brought by these studies, and estimate the effect of decisions by the difference in the impacts of these studies. We release our code and data, which we believe are highly beneficial owing to the scarcity of the dataset on counterfactual studies.

Developed by Humboldt University of Berlin and friends. Flair allows you to apply our state-of-the-art natural language processing (NLP) models to your text, such as named entity recognition (NER), part-of-speech tagging (PoS), special support for biomedical data, sense disambiguation and classification, with support for a rapidly growing number of languages. Flair has simple interfaces that allow you to use and combine different word and document embeddings, including our proposed Flair embeddings, BERT embeddings and ELMo embeddings. Our framework builds directly on PyTorch, making it easy to train your own models and experiment with new approaches using Flair embeddings and classes. New: Most Flair sequence tagging models (named entity recognition, part-of-speech tagging etc.) are now hosted on the HuggingFace model hub!