Considering the one-hidden-layer example above, this corresponds to learning linear predictors over a fixed representation (chosen obliviously and randomly at initialization).

Learning algorithms are often used to make decisions in sequential decision-making environments. In multi-agent settings, the decisions of each agent can affect the utilities/losses of the other agents. Therefore, if an agent is good at anticipating the behavior of the other agents, in particular how they will make decisions in each round as a function of their experience that far, it could try to judiciously make its own decisions over the rounds of the interaction so as to influence the other agents to behave in a way that ultimately benefits its own utility. In this paper, we study repeated two-player games involving two types of agents: a learner, which employs an online learning algorithm to choose its strategy in each round; and an optimizer, which knows the learner's utility function and the learner's online learning algorithm. The optimizer wants to plan ahead to maximize its own utility, while taking into account the learner's behavior.

Considering the one-hidden-layer example above, this corresponds to learning linear predictors over a fixed representation (chosen obliviously and randomly at initialization).

Let H be a hypothesis class with VC dimension d and let 2 (0, 1) . Then there exists a learner Lrn having -adversarial risk " To prove Theorem 3.1, we will use the S SPV and let n 1 / be the sample size. By applying linearity of expectation we get E " 1 t To prove Theorem 3.1, we will need an optimal learner as an input learner for SPV . Theorem 3.1 can now be immediately inferred as a direct application of Lemma A.1 and Theorem A.2 . The impossibility result in Theorem 3.3 extends to randomized learning rules.

While sparse autoencoders (SAEs) have generated significant excitement, a series of negative results have added to skepticism about their usefulness. Here, we establish a conceptual distinction that reconciles competing narratives surrounding SAEs. We argue that while SAEs may be less effective for acting on known concepts, SAEs are powerful tools for discovering unknown concepts. This distinction cleanly separates existing negative and positive results, and suggests several classes of SAE applications. Specifically, we outline use cases for SAEs in (i) ML interpretability, explainability, fairness, auditing, and safety, and (ii) social and health sciences.

The discovery began, as many breakthroughs do, with an observation that didn't quite make sense. In 1948, two French researchers, Paul Mandel and Pierre Métais, published a little-noticed paper in a scientific journal. Working in a laboratory in Strasbourg, they had been cataloguing the chemical contents of blood plasma--that river of life teeming with proteins, sugars, waste, nutrients, and cellular debris. Amid this familiar inventory, they'd spotted an unexpected presence: fragments of DNA drifting freely. The finding defied biological orthodoxy. DNA was thought to remain locked inside the nuclei of cells, and not float around on its own.

Learning algorithms are often used to make decisions in sequential decision-making environments. In multi-agent settings, the decisions of each agent can affect the utilities/losses of the other agents. Therefore, if an agent is good at anticipating the behavior of the other agents, in particular how they will make decisions in each round as a function of their experience that far, it could try to judiciously make its own decisions over the rounds of the interaction so as to influence the other agents to behave in a way that ultimately benefits its own utility. In this paper, we study repeated two-player games involving two types of agents: a learner, which employs an online learning algorithm to choose its strategy in each round; and an optimizer, which knows the learner's utility function and the learner's online learning algorithm. The optimizer wants to plan ahead to maximize its own utility, while taking into account the learner's behavior.