It has been recently observed that neural networks, unlike kernel methods, enjoy a reduced sample complexity when the distribution is isotropic (i.e., when the covariance matrix is the identity). We find that this sensitivity to the data distribution is not exclusive to neural networks, and the same phenomenon can be observed on the class of quadratic classifiers (i.e., the sign of a quadratic polynomial) with a nuclear-norm constraint. We demonstrate this by deriving an upper bound on the Rademacher Complexity that depends on two key quantities: (i) the intrinsic dimension, which is a measure of isotropy, and (ii) the largest eigenvalue of the second moment (covariance) matrix of the distribution. Our result improves the dependence on the dimension over the best previously known bound and precisely quantifies the relation between the sample complexity and the level of isotropy of the distribution.

We study the effect of stochasticity in on-policy policy optimization, and make the following four contributions.

While much attention has been given to the problem of estimating the effect of discrete interventions from observational data, relatively little work has been done in the setting of continuous-valued interventions, such as treatments associated with a dosage parameter. In this paper, we tackle this problem by building on a modification of the generative adversarial networks (GANs) framework. Our model, SCIGAN, is flexible and capable of simultaneously estimating counterfactual outcomes for several different continuous interventions. The key idea is to use a significantly modified GAN model to learn to generate counterfactual outcomes, which can then be used to learn an inference model, using standard supervised methods, capable of estimating these counterfactuals for a new sample. To address the challenges presented by shifting to continuous interventions, we propose a novel architecture for our discriminator - we build a hierarchical discriminator that leverages the structure of the continuous intervention setting. Moreover, we provide theoretical results to support our use of the GAN framework and of the hierarchical discriminator. In the experiments section, we introduce a new semi-synthetic data simulation for use in the continuous intervention setting and demonstrate improvements over the existing benchmark models.

Autonomous agents that control user interfaces to accomplish human tasks are emerging. Leveraging LLMs to power such agents has been of special interest, but unless fine-tuned on human-collected task demonstrations, performance is still relatively low. In this work we study whether fine-tuning alone is a viable approach for building real-world UI control agents. To this end we collect and release a new dataset, AndroidControl, consisting of 15,283 demonstrations of everyday tasks with Android apps. Compared to existing datasets, each AndroidControl task instance includes both high and low-level human-generated instructions, allowing us to explore the level of task complexity an agent can handle.

It has been wide discussed on how to develop algorithms allow large batches, so that one could train neural networks in a distributed environment. The paper investigates the effect of network width on the performance of large-batch training both theoretically and experimentally. The authors claim that with the same number of parameters, it is more likely to train neural networks using proper large batches easily with a wide network architecture. The theoretical support on 2-layers linear/nonlinear networks and multilayer linear networks is also given. The paper is well-written and easy to follow.

Most real-world networks are too large to be measured or studied directly and there is substantial interest in estimating global network properties from smaller sub-samples. One of the most important global properties is the number of vertices/nodes in the network. Estimating the number of vertices in a large network is a major challenge in computer science, epidemiology, demography, and intelligence analysis. In this paper we consider a population random graph G = (V, E) from the stochastic block model (SBM) with K communities/blocks. A sample is obtained by randomly choosing a subset W V and letting G(W) be the induced subgraph in G of the vertices in W. In addition to G(W), we observe the total degree of each sampled vertex and its block membership.

Trained on the voices of popular narrators, the AI's can respond in infinite ways to game scenarios Trained on the voices of popular narrators, the AI's can respond in infinite ways to game scenarios Tractable's tech uses cell phone video to estimate damages from accidents and natural disasters Tractable's tech uses cell phone video to estimate damages from accidents and natural disasters When you check out your fresh produce, the AI recognizes the product, even if it's wrapped in layers When you check out your fresh produce, the AI recognizes the product, even if it's wrapped in layers Submit some samples of your voice and AI will be able to speak exactly as you - what could go wrong? Submit some samples of your voice and AI will be able to speak exactly as you - what could go wrong? Open AI's GPT-3 text generator is released into the wild and generates content that is impermissible Open AI's GPT-3 text generator is released into the wild and generates content that is impermissible

When you check out your fresh produce, the AI recognizes the product, even if it's wrapped in layers When you check out your fresh produce, the AI recognizes the product, even if it's wrapped in layers Submit some samples of your voice and AI will be able to speak exactly as you - what could go wrong? Submit some samples of your voice and AI will be able to speak exactly as you - what could go wrong? Open AI's GPT-3 text generator is released into the wild and generates content that is impermissible Open AI's GPT-3 text generator is released into the wild and generates content that is impermissible "Metacognition" is our ability to reflect on our own certainty - today's AI's are all too confident "Metacognition" is our ability to reflect on our own certainty - today's AI's are all too confident "Turking" for Amazon and its competitors is a lifeline for some, and exploitative grind for others "Turking" for Amazon and its competitors is a lifeline for some, and exploitative grind for others

Perhaps you heard recently about a new algorithm that can drive a car? Or scan a picture and find your face in a crowd? It seems as though every week companies are finding new uses for algorithms that adapt as they encounter new data. Last year Wired quoted an ex-Google employee as saying that "Everything in the company is really driven by machine learning." Machine learning has tremendous potential to transform companies, but in practice it's mostly far more mundane than robot drivers and chefs.

Although researchers have studied planning since the early days of AI, recent developments have revolutionized the field. Furthermore, work on propositional planning is closely related to the algorithms used in the autonomous controller for the National Aeronautics and Space Administration (NASA) Deep Space One spacecraft, launched in October 1998. As a result, our understanding of interleaved planning and execution has advanced as well as the speed with which we can solve classical planning problems. The goal of this survey is to explain these recent advances and suggest new directions for research. Because this article requires minimal AI background (for example, simple logic and basic search algorithms), it's suitable for a wide audience, but my treatment is not exhaustive because I don't have the space to discuss every active topic of planning research.