The paper demonstrates that falsifiability is fundamental to learning. We prove the following theorem for statistical learning and sequential prediction: If a theory is falsifiable then it is learnable -- i.e. admits a strategy that predicts optimally. An analogous result is shown for universal induction.
On-device research index
arXiv research
A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
168,657 papers · 148 categories
Trend · papers per month
48 results for “falsifier”
New framework aims to make neural network explanations more reliable.
problem Current interpretability methods rely on intuition and lack falsifiability.
method Proposes a framework for strongly falsifiable interpretability research.
result Falsifiable interpretability methods can generate meaningful advances in understanding DNNs.
New method falsifies causal graphs using outlier events.
problem Inferring causal relationships from data is hard.
method Falsify candidate causal graphs based on outlier propagation.
result Statistical tests control false positives and have power guarantees.
'Ergodicity economics' is criticized as pseudoscience.
problem Flawed conceptual basis of mainstream economic theory.
method Claims 'ergodicity economics' is more parsimonious and clearer.
result Peters' approach has not produced falsifiable implications.
We information-theoretically reformulate two measures of capacity from statistical learning theory: empirical VC-entropy and empirical Rademacher complexity. We show these capacity measures count the number of hypotheses about a dataset that a learning algorithm falsifies when it finds the classifier in its repertoire …
New method falsifies causal discovery results without ground truth.
problem Evaluation of causal discovery algorithms without ground truth data.
method Detects incompatibilities between causal graphs learned on different subsets of variables.
result Detection of incompatibilities can falsify wrongly inferred causal relations.
New method verifies formulas for causal interventional distributions.
problem Deciding if a given formula correctly identifies an interventional distribution.
method Proposed a falsifier to check if a formula is identifying.
result Falsifier can induce an almost-surely correct verifier for certain models.
A new metric assesses causal graphs using node permutations to detect inconsistencies.
problem Quantifying the goodness of causal graphs and distinguishing them from random graphs.
method Constructing a baseline through node permutations and comparing inconsistencies.
result The proposed metric can distinguish between true and wrong causal graphs.
New neural methods for stable control with provable guarantees.
problem Designing stable control policies for nonlinear systems.
method Neural network Lyapunov functions and a falsifier to guide learning.
result Provable stability of controlled nonlinear systems.
Defensive Generationcs.LG
Efficiently generates models resistant to falsification.
problem Creating models that cannot be disproven by tests.
method Exploits connections between high-dimensional multicalibration and expected variational inequality problems to develop an efficient algorithm.
result First to efficiently produce online outcome indistinguishable generative models resistant to infinite classes of tests.
Machine Learning benefits from prior information and computational power for better performance and understanding.
problem Improper use of Machine Learning methods leads to lack of understanding and performance issues.
method Employing prior information and computational power to solve learning problems, emphasizing interpretability and performance.
result Combining prior information and computational power can lead to better understanding and performance in Machine Learning.
This chapter studies emerging cyber-attacks on reinforcement learning (RL) and introduces a quantitative approach to analyze the vulnerabilities of RL. Focusing on adversarial manipulation on the cost signals, we analyze the performance degradation of TD() and -learning algorithms under the manipulation. For TD($…
Networks of equities in financial marketscond-mat.stat-mech
We review the recent approach of correlation based networks of financial equities. We investigate portfolio of stocks at different time horizons, financial indices and volatility time series and we show that meaningful economic information can be extracted from noise dressed correlation matrices. We show that the metho…
Models can be simple for different reasons: because they yield a simple and computationally efficient interpretation of a generic dataset (e.g. in terms of pairwise dependences) - as in statistical learning - or because they capture the essential ingredients of a specific phenomenon - as e.g. in physics - leading to no…
Theoretical limits show experimental data can falsify but not validate causal estimates from observational studies.
problem Fundamental limits on validating causal estimates using experimental data in observational studies.
method Impossible inference framework, Gaussian Process based approach.
result Experimental data can falsify but not validate causal estimates from observational studies.
Faster Tsetlin Machines use clause indexing to speed inference and learning.
problem Overfitting and slow inference in Tsetlin Machines.
method Introduced a look-up table that indexes clauses based on feature falsification, enabling faster evaluation of clauses.
result Up to 15 times faster classification and three times faster learning on MNIST and Fashion-MNIST.
CorrGAN: Sampling Realistic Financial Correlation Matrices Using Generative Adversarial Networksq-fin.ST
We propose a novel approach for sampling realistic financial correlation matrices. This approach is based on generative adversarial networks. Experiments demonstrate that generative adversarial networks are able to recover most of the known stylized facts about empirical correlation matrices estimated on asset returns.…
New formulas for minimal surfaces with specific end conditions.
problem Existence and explicit formulas for minimal surfaces with embedded planar ends.
method Provided new explicit formulas for genus 0 minimal surfaces in R^3 with 2k+1 embedded planar ends.
result Existence and explicit formulas for minimal surfaces with 2k+1 embedded planar ends for all k ≥ 4.
Finding regions for which there is higher controversy among different classifiers is insightful with regards to the domain and our models. Such evaluation can falsify assumptions, assert some, or also, bring to the attention unknown phenomena. The present work describes an algorithm, which is based on the Exceptional M…
We prove that a four-dimensional Lorentzian manifold that is curvature homogeneous of order 3, or CH_3 for short, is necessarily locally homogeneous. We also exhibit and classify four-dimensional Lorentzian, CH_2 manifolds that are not homogeneous. The resulting metrics belong to the class of null electromagnetic radia…
Causal Bayesian networks interpret actions as interventions to connect models to real-world outcomes.
problem Connecting causal model predictions to real-world outcomes.
method Formal framework to interpret actions as interventions and prove impossibility results.
result No non-circular interpretation exists that satisfies natural desiderata without violating some.
We find a remarkable time persistence of various proxies for the kurtosis (p-kurtosis) of the intraday returns distribution for the S&P500 index and this permits a significant measure of their evolution from 1983 to 2004. There appears a long time scale dramatic variation of the p-kurtosis uncorrelated with the variati…
A geometric theory explains loss functions for robust representation learning.
problem Treats robustness, domain adaptation, and sensor drift as separate literatures.
method Estimates covariance Sigma_task and uses it to pin Jacobian penalties.
result Proves optimality and necessity of range coverage for penalty matrices.
Audit financial machine learning workflows to detect spurious predictability.
problem Spurious predictability in financial machine learning models.
method Falsification audit testing predictive workflows against synthetic environments.
result Many apparent financial predictions are artifacts, not genuine.
New insights into training deep networks with low rank layers.
problem Efficiency in training deep neural networks.
method Analysis of techniques for training in low rank space.
result Falsified common beliefs in training deep networks.
Assumption-lean falsification tests of rate double-robustness of double-machine-learning estimatorsstat.ME
Tests validity of DML estimators without assumptions.
problem Validating DML estimators without making assumptions.
method Develops tests to falsify assumptions for DML estimators.
result Falsifies assumptions for DML estimators with non-trivial power.
Accounting fraud is a global concern representing a significant threat to the financial system stability due to the resulting diminishing of the market confidence and trust of regulatory authorities. Several tricks can be used to commit accounting fraud, hence the need for non-static regulatory interventions that take …
Resilient Learning-Based Control for Synchronization of Passive Multi-Agent Systems under Attackeess.SY
In this paper, we show synchronization for a group of output passive agents that communicate with each other according to an underlying communication graph to achieve a common goal. We propose a distributed event-triggered control framework that will guarantee synchronization and considerably decrease the required comm…
Study finds optimal learning rate schedules for sub-100M quantization-aware training across bit-widths.
problem Optimal learning rate schedules for quantization-aware training depend on bit-width.
method Factorial grid testing over bit-width, warmdown fraction, LR magnitude, model size, and seed.
result INT6 QAT requires a different schedule than higher-precision training, falsifying the primary hypothesis.
LLM agents discover cryptocurrency factors under reproducible constraints.
problem Flexibility of LLM agents in empirical discovery leads to uncontrolled search.
method Sequential hypothesis search with fixed data splits and portfolio tests.
result Ridge-combined portfolio achieves 44.55% annualized return in out-of-sample period.
Algorithm detects unmeasured confounding in observational data.
problem Estimating treatment effects in observational studies with untestable conditions.
method Two-stage procedure that detects dependencies between causal mechanisms.
result Algorithm efficiently detects confounding on simulated and semi-synthetic data.
Instrumented data enables causal scientific machine learning
problem Insufficient data for causal scientific machine learning
method Instrumented data with explicit model, uncertainty, and counterfactuals
result Supports causal interventions through Pearl's do-operator
In the domain of the so called Econophysics some attempts already have been made for applying the theory of Thermodynamics and Statistical Mechanics to economics and financial markets. In this paper a similar approach is made from a different perspective, trying to model the limit order book and price formation process…
Foreign policy analysis has been struggling to find ways to measure policy preferences and paradigm shifts in international political systems. This paper presents a novel, potential solution to this challenge, through the application of a neural word embedding (Word2vec) model on a dataset featuring speeches by heads o…
This paper considers the problem of detection in distributed networks in the presence of data falsification (Byzantine) attacks. Detection approaches considered in the paper are based on fully distributed consensus algorithms, where all of the nodes exchange information only with their neighbors in the absence of a fus…
Triangulation filters spurious circuits in multilingual models.
problem Unreliable explanations of multilingual models across languages.
method Formalizes reference families and introduces triangulation as a causal acceptance rule.
result Triangulation provides a falsifiable standard for mechanistic claims.
New analysis shows surprising results on adaptation speed of causal models.
problem Investigate the adaptation speed of causal models under interventions.
method Use convergence rates from stochastic optimization to measure adaptation speed.
result Surprising findings: anticausal model can be faster than causal model under certain conditions.
The paper benchmarks OS with RCTs, accounting for right-censoring.
problem Benchmarking observational studies with experimental data under censoring.
method Two cases: independent and dependent censoring. Censoring-doubly-robust signal for CATE.
result Effectiveness of censoring-aware tests verified via experiments and real data.
Hyperfitting improves LLM generation quality by enhancing diversity, contrary to simple temperature scaling.
problem Improving open-ended generation quality of LLMs with minimal fine-tuning effort.
method Demonstrates that hyperfitting, a phenomenon where LLMs are fine-tuned to near-zero training loss, enhances generation quality and mitigates repetition.
result Hyperfitting is distinct from temperature scaling and involves a dynamic, context-dependent rank reordering mechanism in the final transformer block.
New analysis shows LLMs don't follow Bayesian inference in ICL.
problem Does in-context learning in LLMs follow Bayesian inference?
method Analyzes ICL through the martingale property, a requirement for Bayesian inference.
result Violations of the martingale property show LLMs don't follow Bayesian inference.
The recent advances in sensor technologies and smart devices enable the collaborative collection of a sheer volume of data from multiple information sources. As a promising tool to efficiently extract useful information from such big data, machine learning has been pushed to the forefront and seen great success in a wi…
Evidence shows that in a significant number of cases the current methods of research do not allow for reproducible and falsifiable procedures of scientific investigation. As a consequence, the majority of critical decisions at all levels, from personal investment choices to overreaching global policies, rely on some va…
Is cognition a collection of loosely connected functions tuned to different tasks, or can there be a general learning algorithm? If such an hypothetical general algorithm did exist, tuned to our world, could it adapt seamlessly to a world with different laws of nature? We consider the theory that predictive coding is s…
We use an adversarial expert based online learning algorithm to learn the optimal parameters required to maximise wealth trading zero-cost portfolio strategies. The learning algorithm is used to determine the relative population dynamics of technical trading strategies that can survive historical back-testing as well a…
Unification of field theory and maximum entropy methods for learning probability densitiesphysics.data-an
The need to estimate smooth probability distributions (a.k.a. probability densities) from finite sampled data is ubiquitous in science. Many approaches to this problem have been described, but none is yet regarded as providing a definitive solution. Maximum entropy estimation and Bayesian field theory are two such appr…
For many causal effect parameters of interest, doubly robust machine learning (DRML) estimators are the state-of-the-art, incorporating the good prediction performance of machine learning; the decreased bias of doubly robust estimators; and the analytic tractability and bias reduction of sample splitting wi…
CJE calibrates cheap LLM judges against an oracle, achieving high accuracy at a fraction of the cost.
problem Inexpensive LLM judges can produce biased rankings, leading to unreliable outcomes.
method CJE uses a small oracle to calibrate cheap scores, then evaluates at scale with valid uncertainty.
result CJE achieves 99% pairwise ranking accuracy at 14x lower cost compared to a 16x oracle/judge cost ratio.
There are (at least) three approaches to quantifying information. The first, algorithmic information or Kolmogorov complexity, takes events as strings and, given a universal Turing machine, quantifies the information content of a string as the length of the shortest program producing it. The second, Shannon information…