Publications associées à Impact evaluation

SIMPLE HIERARCHICAL PLANNING WITH DIFFUSION

Diffusion-based generative methods have proven effective in modeling trajectories with offline datasets. However, they often face computational challenges and can falter in generalization, especially in capturing temporal abstractions for long- horizon tas ...

2024

Discussion of ‘Experimental Evaluation of Algorithm-Assisted Human Decision-Making: Application to Pretrial Public Safety Assessment’, part 2

Mats Julius Stensrud, Aaron Leor Sarvet

Imai et al.(IJGHS) have conducted a timely experiment on evaluating a decision support algorithm. However, we are concerned by their choice of estimands which, even if they appear plausible at first, rely on notions and assumptions for which we cannot ever ...

2023

Sediment continuity and augmentation measures

Giovanni De Cesare, Robin Schroff

Impaired sediment transport can have numerous adverse impacts on the eco-morphodynamics of the riverscape. If well designed, sediment augmentation measures present a promising mitigation approach at different scales. This chapter focuses on flume experim ...

2023

Evaluation and digitalization: what are the key challenges for evaluation processes and evaluators?

Omar Ballester Gonzalez

Public policy evaluation has traditionally been conducted in a context constrained by legal, institutional and political red tape. The arrival of digitalisation adds a new layer to public policy evaluation processes. Assessing the quality of technologies, ...

Edward Elgar Publishing2023

RDS: FPGA Routing Delay Sensors for Effective Remote Power Analysis Attacks

Mirjana Stojilovic, Ognjen Glamocanin, David Spielmann

State-of-the-art sensors for measuring FPGA voltage fluctuations are time-to-digital converters (TDCs). They allow detecting voltage fluctuations in the order of a few nanoseconds. The key building component of a TDC is a delay line, typically implemented ...

2023

DISCO: Distilling Counterfactuals with Large Language Models

Antoine Bosselut, Zeming Chen, Qiyue Gao

Models trained with counterfactually augmented data learn representations of the causal structure of tasks, enabling robust generalization. However, high-quality counterfactual data is scarce for most tasks and not easily generated at scale. When crowdsour ...

Assoc Computational Linguistics-Acl2023

Towards Novel Evaluation Methods for Social Dialog Systems

Ekaterina Svikhnushina

Language has shaped human evolution and led to the desire to endow machines with language abilities. Recent advancements in natural language processing enable us to achieve this breakthrough in human-machine interaction. However, introducing conversational ...

EPFL2023

Serab: A Multi-Lingual Benchmark For Speech Emotion Recognition

Pierre Anton Beckmann, Milos Cernak

Recent developments in speech emotion recognition (SER) often leverage deep neural networks (DNNs). Comparing and benchmarking different DNN models can often be tedious due to the use of different datasets and evaluation protocols. To facilitate the proces ...

IEEE2022

Efficient and Effective Multi-Modal Queries Through Heterogeneous Network Embedding

Karl Aberer, Quoc Viet Hung Nguyen, Thanh Trung Huynh, Thành Tâm Nguyên, Chi Thang Duong

The heterogeneity of today's Web sources requires information retrieval (IR) systems to handle multi-modal queries. Such queries define a user's information needs by different data modalities, such as keywords, hashtags, user profiles, and other media. Rec ...

IEEE COMPUTER SOC2022

Memory-Based Model Editing at Scale

Antoine Bosselut

Even the largest neural networks make errors, and once-correct predictions can become invalid as the world changes. Model editors make local updates to the behavior of base (pre-trained) models to inject updated knowledge or correct undesirable behaviors. ...

JMLR-JOURNAL MACHINE LEARNING RESEARCH2022