We introduce a method to provide vectorial representations of visual classification tasks which can be used to reason about the nature of those tasks and their relations. Given a dataset with ground-truth labels and a loss function defined over those labels, we process images through a "probe network" and compute an embedding based on estimates of the Fisher information matrix associated with the probe network parameters. This provides a fixed-dimensional embedding of the task that is independent of details such as the number of classes and does not require any understanding of the class label semantics. We demonstrate that this embedding is capable of predicting task similarities that match our intuition about semantic and taxonomic relations between different visual tasks (e.g., tasks based on classifying different types of plants are similar) We also demonstrate the practical value of this framework for the meta-task of selecting a pre-trained feature extractor for a new task. We present a simple meta-learning framework for learning a metric on embeddings that is capable of predicting which feature extractors will perform well. Selecting a feature extractor with task embedding obtains a performance close to the best available feature extractor, while costing substantially less than exhaustively training and evaluating on all available feature extractors.
How can we help a forgetful learner learn multiple concepts within a limited time frame? For long-term learning, it is crucial to devise teaching strategies that leverage the underlying forgetting mechanisms of the learner. In this paper, we cast the problem of adaptively teaching a forgetful learner as a novel discrete optimization problem, where we seek to optimize a natural objective function that characterizes the learner's expected performance throughout the teaching session. We then propose a simple greedy teaching strategy and derive strong performance guarantees based on two intuitive data-dependent properties, which capture the degree of diminishing returns of teaching each concept. We show that, given some assumptions about the learner's memory model, one can efficiently compute the performance bounds. Furthermore, we identify parameter settings of the memory model where the greedy strategy is guaranteed to achieve high performance. We demonstrate the effectiveness of our algorithm using extensive simulations along with user studies in two concrete applications, namely (i) an educational app for online vocabulary teaching and (ii) an app for teaching novices how to recognize animal species from images.
We address the problem of 3D human pose estimation from 2D input images using only weakly supervised training data. Despite showing considerable success for 2D pose estimation, the application of supervised machine learning to 3D pose estimation in real world images is currently hampered by the lack of varied training images with corresponding 3D poses. Most existing 3D pose estimation algorithms train on data that has either been collected in carefully controlled studio settings or has been generated synthetically. Instead, we take a different approach, and propose a 3D human pose estimation algorithm that only requires relative estimates of depth at training time. Such training signal, although noisy, can be easily collected from crowd annotators, and is of sufficient quality for enabling successful training and evaluation of 3D pose algorithms. Our results are competitive with fully supervised regression based approaches on the Human3.6M dataset, despite using significantly weaker training data. Our proposed algorithm opens the door to using existing widespread 2D datasets for 3D pose estimation by allowing fine-tuning with noisy relative constraints, resulting in more accurate 3D poses.
It is desirable for detection and classification algorithms to generalize to unfamiliar environments, but suitable benchmarks for quantitatively studying this phenomenon are not yet available. We present a dataset designed to measure recognition generalization to novel environments. The images in our dataset are harvested from twenty camera traps deployed to monitor animal populations. Camera traps are fixed at one location, hence the background changes little across images; capture is triggered automatically, hence there is no human bias. The challenge is learning recognition in a handful of locations, and generalizing animal detection and classification to new locations where no training data is available. In our experiments state-of-the-art algorithms show excellent performance when tested at the same location where they were trained. However, we find that generalization to new locations is poor, especially for classification systems.
In real-world applications of education, an effective teacher adaptively chooses the next example to teach based on the learner's current state. However, most existing work in algorithmic machine teaching focuses on the batch setting, where adaptivity plays no role. In this paper, we study the case of teaching consistent, version space learners in an interactive setting. At any time step, the teacher provides an example, the learner performs an update, and the teacher observes the learner's new state. We highlight that adaptivity does not speed up the teaching process when considering existing models of version space learners, such as "worst-case" (the learner picks the next hypothesis randomly from the version space) and "preference-based" (the learner picks hypothesis according to some global preference). Inspired by human teaching, we propose a new model where the learner picks hypotheses according to some local preference defined by the current hypothesis. We show that our model exhibits several desirable properties, e.g., adaptivity plays a key role, and the learner's transitions over hypotheses are smooth/interpretable. We develop efficient teaching algorithms and demonstrate our results via simulation and user studies.
Existing image classification datasets used in computer vision tend to have a uniform distribution of images across object categories. In contrast, the natural world is heavily imbalanced, as some species are more abundant and easier to photograph than others. To encourage further progress in challenging real world conditions we present the iNaturalist species classification and detection dataset, consisting of 859,000 images from over 5,000 different species of plants and animals. It features visually similar species, captured in a wide variety of situations, from all over the world. Images were collected with different camera types, have varying image quality, feature a large class imbalance, and have been verified by multiple citizen scientists. We discuss the collection of the dataset and present extensive baseline experiments using state-of-the-art computer vision classification and detection models. Results show that current non-ensemble based methods achieve only 67% top one classification accuracy, illustrating the difficulty of the dataset. Specifically, we observe poor results for classes with small numbers of training examples suggesting more attention is needed in low-shot learning.
We present and evaluate the Fast (conditional) Independence Test (FIT) -- a nonparametric conditional independence test. The test is based on the idea that when $P(X \mid Y, Z) = P(X \mid Y)$, $Z$ is not useful as a feature to predict $X$, as long as $Y$ is also a regressor. On the contrary, if $P(X \mid Y, Z) \neq P(X \mid Y)$, $Z$ might improve prediction results. FIT applies to thousand-dimensional random variables with a hundred thousand samples in a fraction of the time required by alternative methods. We provide an extensive evaluation that compares FIT to six extant nonparametric independence tests. The evaluation shows that FIT has low probability of making both Type I and Type II errors compared to other tests, especially as the number of available samples grows. Our implementation of FIT is publicly available.
Low dimensional embeddings that capture the main variations of interest in collections of data are important for many applications. One way to construct these embeddings is to acquire estimates of similarity from the crowd. However, similarity is a multi-dimensional concept that varies from individual to individual. Existing models for learning embeddings from the crowd typically make simplifying assumptions such as all individuals estimate similarity using the same criteria, the list of criteria is known in advance, or that the crowd workers are not influenced by the data that they see. To overcome these limitations we introduce Context Embedding Networks (CENs). In addition to learning interpretable embeddings from images, CENs also model worker biases for different attributes along with the visual context i.e. the visual attributes highlighted by a set of images. Experiments on two noisy crowd annotated datasets show that modeling both worker bias and visual context results in more interpretable embeddings compared to existing approaches.
We study the problem of computer-assisted teaching with explanations. Conventional approaches for machine teaching typically only provide feedback at the instance level e.g., the category or label of the instance. However, it is intuitive that clear explanations from a knowledgeable teacher can significantly improve a student's ability to learn a new concept. To address these existing limitations, we propose a teaching framework that provides interpretable explanations as feedback and models how the learner incorporates this additional information. In the case of images, we show that we can automatically generate explanations that highlight the parts of the image that are responsible for the class label. Experiments on human learners illustrate that, on average, participants achieve better test set performance on challenging categorization tasks when taught with our interpretable approach compared to existing methods.
We propose and systematically evaluate three strategies for training dynamically-routed artificial neural networks: graphs of learned transformations through which different input signals may take different paths. Though some approaches have advantages over others, the resulting networks are often qualitatively similar. We find that, in dynamically-routed networks trained to classify images, layers and branches become specialized to process distinct categories of images. Additionally, given a fixed computational budget, dynamically-routed networks tend to perform better than comparable statically-routed networks.