Label corruption, where training samples have incorrect labels, can significantly degrade the performance of machine learning models. This corruption often arises from non-expert labeling or adversarial attacks. Acquiring large, perfectly labeled datasets is costly, and retraining large models from scratch when a clean dataset becomes available is computationally expensive. To address this challenge, we propose Post-Training Correction, a new paradigm that adjusts model parameters after initial training to mitigate label noise, eliminating the need for retraining. We introduce Verifix, a novel Singular Value Decomposition (SVD) based algorithm that leverages a small, verified dataset to correct the model weights using a single update. Verifix uses SVD to estimate a Clean Activation Space and then projects the model's weights onto this space to suppress activations corresponding to corrupted data. We demonstrate Verifix's effectiveness on both synthetic and real-world label noise. Experiments on the CIFAR dataset with 25% synthetic corruption show 7.36% generalization improvements on average. Additionally, we observe generalization improvements of up to 2.63% on naturally corrupted datasets like WebVision1.0 and Clothing1M.

Causal representation learning aims at identifying high-level causal variables from perceptual data. Most methods assume that all latent causal variables are captured in the high-dimensional observations. We instead consider a partially observed setting, in which each measurement only provides information about a subset of the underlying causal state. Prior work has studied this setting with multiple domains or views, each depending on a fixed subset of latents. Here, we focus on learning from unpaired observations from a dataset with an instance-dependent partial observability pattern. Our main contribution is to establish two identifiability results for this setting: one for linear mixing functions without parametric assumptions on the underlying causal model, and one for piecewise linear mixing functions with Gaussian latent causal variables. Based on these insights, we propose two methods for estimating the underlying causal variables by enforcing sparsity in the inferred representation. Experiments on different simulated datasets and established benchmarks highlight the effectiveness of our approach in recovering the ground-truth latents.

This model is known as the single-index model in statistics, and, among other things, it represents a significant generalization of one-bit compressed sensing.

We consider the following multi-component sparse PCA problem: given a set of data points, we seek to extract a small number of sparse components with disjoint supports that jointly capture the maximum possible variance. Such components can be computed one by one, repeatedly solving the single-component problem and deflating the input data matrix, but this greedy procedure is suboptimal. We present a novel algorithm for sparse PCA that jointly optimizes multiple disjoint components. The extracted features capture variance that lies within a multiplicative factor arbitrarily close to 1 from the optimal. Our algorithm is combinatorial and computes the desired components by solving multiple instances of the bipartite maximum weight matching problem. Its complexity grows as a low order polynomial in the ambient dimension of the input data, but exponentially in its rank. However, it can be effectively applied on a low-dimensional sketch of the input data. We evaluate our algorithm on real datasets and empirically demonstrate that in many cases it outperforms existing, deflation-based approaches.

Google is restricting its Gemini AI chatbot from answering election-related questions in countries where voting is taking place this year, limiting users from receiving information about candidates, political parties and other elements of politics. "Out of an abundance of caution on such an important topic, we have begun to roll out restrictions on the types of election-related queries for which Gemini will return responses," Google's India team stated on the company's site. The company initially announced its plans for limiting election-related queries in a blog post last December, according to a Google spokesperson, and made a similar announcement regarding European parliamentary elections in February. Google's post on Tuesday pertained to India's upcoming election, while TechCrunch reported that Google confirmed it is rolling out the changes globally. When asked questions such as "tell me about President Biden" or "who is Donald Trump," Gemini now replies: "I'm still learning how to answer this question. In the meantime, try Google search," or a similarly evasive answer.

Solving real world problems with embedded neural networks requires both training algorithms that achieve high performance and compatible hardware that runs in real time while remaining energy efficient. For the former, deep learning using backpropagation has recently achieved a string of successes across many domains and datasets. For the latter, neuromorphic chips that run spiking neural networks have recently achieved unprecedented energy efficiency. To bring these two advances together, we must first resolve the incompatibility between backpropagation, which uses continuous-output neurons and synaptic weights, and neuromorphic designs, which employ spiking neurons and discrete synapses. Our approach is to treat spikes and discrete synapses as continuous probabilities, which allows training the network using standard backpropagation. The trained network naturally maps to neuromorphic hardware by sampling the probabilities to create one or more networks, which are merged using ensemble averaging. To demonstrate, we trained a sparsely connected network that runs on the TrueNorth chip using the MNIST dataset. With a high performance network (ensemble of 64), we achieve 99.42% accuracy at 108 µJ per image, and with a high efficiency network (ensemble of 1) we achieve 92.7% accuracy at 0.268 µJ per image.

Gemini, Google's AI chatbot, won't answer questions about India's upcoming national elections, the company wrote in a blog post today. "Out of an abundance of caution on such an important topic, we have begun to roll out restrictions on the types of election-related queries for which Gemini will return responses," the company wrote. The restrictions are similar to the ones Google announced in December ahead of global elections in the US and the EU. "As we shared last December, in preparation for the many elections happening around the world in 2024 and out of an abundance of caution, we're restricting the types of election-related queries for which Gemini will return responses," a Google spokesperson wrote to Engadget. The guardrails are already in place in the US. When I asked Gemini for interesting facts about the 2024 US presidential election, it replied, "I'm still learning how to answer this question.

Ukraine launched one of its most severe attacks into Russia on Tuesday – destroying critical energy infrastructure. Meanwhile, Russia claims it is doing everything necessary to thwart new attacks – including the invasion of anti-Kremlin troops threatening to disrupt the upcoming presidential election.

Gibbs sampling is a Markov Chain Monte Carlo sampling technique that iteratively samples variables from their conditional distributions. There are two common scan orders for the variables: random scan and systematic scan. Due to the benefits of locality in hardware, systematic scan is commonly used, even though most statistical guarantees are only for random scan. While it has been conjectured that the mixing times of random scan and systematic scan do not differ by more than a logarithmic factor, we show by counterexample that this is not the case, and we prove that that the mixing times do not differ by more than a polynomial factor under mild conditions. To prove these relative bounds, we introduce a method of augmenting the state space to study systematic scan using conductance.

In this work, we propose an infinite restricted Boltzmann machine (RBM), whose maximum likelihood estimation (MLE) corresponds to a constrained convex optimization. We consider the Frank-Wolfe algorithm to solve the program, which provides a sparse solution that can be interpreted as inserting a hidden unit at each iteration, so that the optimization process takes the form of a sequence of finite models of increasing complexity. As a side benefit, this can be used to easily and efficiently identify an appropriate number of hidden units during the optimization. The resulting model can also be used as an initialization for typical state-of-the-art RBM training algorithms such as contrastive divergence, leading to models with consistently higher test likelihood than random initialization.